Figures
Abstract
Background
Prevalence of postnatal depression (PND) is high (Western countries, 10–15%; Japan, 17%). PND can cause parenting impairment and affect family health (e.g. child behaviors, cognitive development and physical health). This study aimed to reveal the risk factors of PND during the pregnancy period in a Japanese sample, and to identify the psychosocial risk factors of PND that should be appended to existing obstetric interview sheets. A cohort study with a Japanese sample was conducted.
Methods
All 14 obstetrics hospitals in the Setagaya ward, Tokyo, Japan, participated in this study. Pregnant women who booked their delivery between December 2012 and May 2013 were enrolled. Data used for this study were collected at 20 weeks gestation, a few days and one month postnatal. The questionnaires consisted of psychosocial factors and the Edinburgh Postnatal Depression Scale (EPDS). To identify PND risk factors, multivariate analyses were performed.
Results
A total of 1,775 women participated in this study. Eventually, the data of 1,133 women were used for the multivariate analyses. The demonstrated significant risk factors include EPDS score, primipara, “a perceived lack of family cohesion”, “current physical illness treatment” and “current psychiatric illness treatment”.
Conclusion
This study highlights the importance of mental health screening using psychological measures during the pregnancy period. In addition, family environment, parity, physical and psychiatric illness should be paid attention by professionals in maternal and child health. The results also suggest that mothers’ feelings of developing their families should be supported.
Citation: Tachibana Y, Koizumi T, Takehara K, Kakee N, Tsujii H, Mori R, et al. (2015) Antenatal Risk Factors of Postpartum Depression at 20 Weeks Gestation in a Japanese Sample: Psychosocial Perspectives from a Cohort Study in Tokyo. PLoS ONE 10(12): e0142410. https://doi.org/10.1371/journal.pone.0142410
Editor: Kenji Hashimoto, Chiba University Center for Forensic Mental Health, JAPAN
Received: June 12, 2015; Accepted: October 20, 2015; Published: December 1, 2015
Copyright: © 2015 Tachibana et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper.
Funding: This study was conducted under the title, “Research on the assessment of the mental health condition of women during the prenatal and postnatal periods, and appropriate interventions”, supported by the Health and Labor Sciences Research Grant on Children, Youth and Families (H24-Jisedai-Shitei-006), and the title “Research on the promotion for networks cooperating with hospitals, public health, and welfare to support women with depression in the perinatal periods” (H25-Seisin-Wakate-013) by the Health and Labor Science Research Grant on Psychiatry and Neurological Disease and Mental Health, provided by the Ministry of Health, Labor and Welfare of Japan, and the title “Research on child abuse prevention and follow-up for families at high risk of child abuse” by Research Development Grant for Child Health and Development from National Center for Child Health and Development (H26-48).
Competing interests: The authors have declared that no competing interests exist.
Introduction
The prevalence of postnatal depression (PND) is high. In Western countries, the prevalence of postnatal women experiencing a major depressive episode during this period is 10–15% [1]. In Japan, it is estimated to be at 17% [2]. PND can cause parenting impairment [3–5] and have negative effect on child behaviors [6], cognitive development [7] and physical health [8].
Obstetricians and midwives can serve as gatekeepers for detecting PND, as they see the women during pregnancy and the postnatal period. Attention to the risk factors of PND can lead to early detections and interventions for affected women. Previous studies performed in the U.S. and Europe revealed that depressed mood or anxiety during pregnancy, the level of social support, life events and psychiatric history, including previous experience of depression, poor marital relationship and low social status, are all important risk factors of depression in the postnatal period [1, 9, 10].
In Asia, two studies have suggested the sex of the newborn to be a risk factor of PND because of a societal preference for male offspring. In their studies, the other risk factors are consistent with previous studies; i.e. antenatal psychiatric morbidity, economic deprivation, low education, and marital disharmony and protective factors; education, support from extended family members, and employment [11, 12]. There are some studies that reported the antenatal risk factors for PND in Japan. In the study by Kitamura et al., in which 290 Japanese women expecting their first baby (among 1,159 women attending the antenatal clinic of the five university hospitals) were exmined, it was stated that PND is characterized by poor accommodation, dissatisfaction with sex of the newborn baby and with the emotional undermining [13]. Meanwhile, the Osaka Maternal and Child Health Study cohort study of 627 pregnant women conducted in Neyagawa city in Osaka revealed that job, especially for those holding a professional or technical full-time job, is significantly associated with a reduced risk of PND [14]. On the other hand, household income or maternal and paternal educational levels are not. Kokubu et al. reported that anxiety during pregnancy among 99 women attending four antenatal clinics predicts PND, and the effect of negative attitudes towards pregnancy on PND is possibly mediated by bonding failure [15]. Morikawa et al. revealed that social support and depressive symptoms among 877 women during pregnancy affect PND [16]. In view of these studies, we hypothesized that a large community-based cohort study that includes all the hospitals in one area conducted during the perinatal periods can add strong evidence of risk factors for PND. In addition, we think that perinatal professionals should pay attention to women at high risk for PND during the antenatal period. In Japan, home visits by public health nurses are conducted for almost all families with a newborn. The visits, done between one and four months after delivery, serve as an important aspect of the health policy in Japan to prevent child abuse and maltreatment, as well as to support child rearing and family mental health. At present, mental health screening during pregnancy is not a routine procedure in Japanese obstetric outpatient clinics. Although interview sheets about psychosocial aspects are commonly used, they are not considered to be reliable from the viewpoint of mental health practices. Psychosocial risk factors of PND in the pregnancy periods are useful for the screening at obstetric outpatient clinics to perform early interventions for women with high-risks of mental health.
The purpose of this study was to demonstrate the important aspects that should be paid attention to during the pregnancy period to predict PND. We conducted a cohort study on antenatal and postnatal mental health to determine the risk factors of PND in a Japanese sample. Using the cohort study data, we investigated the important psychosocial risk factors of PND.
Methods
Overall study design
This was a cohort study. The project commenced in September 2012, with the cohort recruited from pregnant women in their second trimester in the Setagaya ward of Tokyo. Women who gave informed consent were asked to answer a survey at 20 weeks gestation and at five time points postnatal; the first few days, two weeks, one month, two months, and three months postnatal. At the first survey, subsequent questionnaires were distributed to the women unless they indicated their intention to withdraw from the study, or if they did not complete the questionnaire at the time point. Data for the survey at 20 weeks gestation, the first few days and one month after delivery were collected via the self-administered paper questionnaires or electronically using MOMONGA (Xware Corp, Tokyo, Japan), an iPad2 questionnaire application, at prenatal and postnatal check-ups, as well as during admission after delivery. One month postnatal questionnaires were distributed by care providers and returned to them directly at the hospital. For this study, we used the data at 20 weeks gestation (T1), a few days postnatal (T2), and one month postnatal (T3). Recruitment of the participants was conducted between December 2012 and May 2013. The flow of participants is shown in Fig 1.
Participants
Participants are pregnant women who had been followed up at all the obstetrics outpatient clinics in the Setagaya ward (14 obstetrics hospitals), were under 20 weeks of gestation, and made appointment to deliver at those clinics. Exclusion criteria included women with intellectual disability, learning disorders or insufficient level of Japanese to complete the questionnaires, and those who did not deliver at the obstetrics hospitals in the Setagaya ward. In the event that a participant gave birth to a stillborn baby, the participant was withdrawn from the study. The research staff was updated with the situation by the perinatal staff. In turn, the research staff deleted the details of the participants from the delivery list for the self-administered paper questionnaires. Obstetricians, midwives and nurses supported the women in the same way as they would for all other cases. If a participant had severe mental health problems, medical care was administered by the perinatal staff. Subsequently, the participant was referred to another psychiatric clinic or hospital according to normal protocol.
Recruitment
Two obstetric hospital networks in the Setagaya ward that were involved in the Japan Association of Obstetricians and Gynecologists were approached. Following presentations at regional network meetings, we met with the obstetric hospital directors who were interested to discuss the project details. A hospital would be accepted into the study if the obstetric doctors and nurses indicated support together with the director agreeing to the provision of necessary data. Eventually, all 14 obstetric hospitals in the Setagaya ward were accepted to be part of this cohort study. Participants were recruited via all the hospitals.
Measures
The T1 questionnaire consisted of psychosocial questions and the Japanese version [17] of the Edinburgh Postnatal Depression Scale (EPDS) [18]. T1 was developed based on known predictors of PND and the standard obstetrical interview sheets in Japan. In addition, all the risk factors which were revealed by Robertson [10] and O’Hara [1] (mentioned in the introduction) were included. Low social status was defined by annual household income of less than two million yen. This income thresholds is a representative definition of the working poor in Japan [19]. Table 1 shows the demographic information collected at T1, which includes age, employment, education, income, plurality, parity, wanted pregnancy, psychiatric illness history, reproductive treatment history, type of pregnancy, delivery week.
The T2 data was about parity, and delivery week and method of birth. The T3 data was solely the EPDS.
Data preparation
Data collected were entered into an electronic database. All manually processed questionnaires were double-checked for data quality. All measurements were examined for their ranges, distributions, means, standard deviations, outliers, and logical errors.
Privacy protection
Databases needed for answering specific research questions were centrally built from databases concerning different time points of the study. All information enabling identification of the participants, with the exception the identification number of each participant, was erased from these databases.
Statistical Analyses
The participants were classified as high-risk of PND or not using the cut-off score of the Japanese version of the EPDS (at 8/9) [17]. To explore the postnatal risk factors in preterm periods and soon after delivery, demographic, sociological, psychological, and psychiatric variables were included the preliminary bivariate analyses. Only “parity” among the T2 data was included in the analyses (“delivery week” and “method of birth” were not included in the analyses). All of the analyses were performed using the maximum samples with excluding missing values. We tested intergroup differences in categorical and continuous variables with the chi-square tests and Student’s t test for unpaired data, respectively. When a variable was significantly correlated to other similar independent variables), we selected variables for examining the determinants of PND risk factors. All p values reported were two-tailed. The variables which showed statistical significances were compared to the risk factors in the previous studies of Robertson and O’Hara [1, 10] to confirm if the variables covered were risk factors or not.
Analysis 1.
To identify the risk factors of PND, variables with p value of less than 0.05 in the bivariate analysis were entered into a multivariate logistic regression model as independent variables and the classified two groups with the EPDS cut-off score as a dependent variable. The variables which showed p values of 0.05 or less were considered indicative of statistically significant risk factor of PND.
Analysis 2.
To identify the significant psychosocial factors for interview sheets besides psychometric measurement of depression and anxiety, we performed a logistic regression analysis using the same variables besides the EPDS for independent variables and the classified two groups with the EPDS cut-off score as a dependent variable. All data analyses were performed with SPSS version 21.0 J for Windows (SPSS Inc., Tokyo, Japan).
Multicollinearities of the logistic regression models of Analyses 1 and 2 were estimated. They were performed in the coefficiency estimates of the linear regression in which the same dependent variable and independent variables of the logistic regression analysis were used. We judged the presence of multicollinearity if tolerance values were less than 0.4 and the variance inflation factor (VIF) was greater than 2.5 using Allison’s criteria [20].
Subanalysis
The antenatal risk factor of PND may also have a strong relationship with antenatal depression. Since antenatal depression was measured by EPDS, we performed subanalyses classifying the participants into two groups: women at high risk of depression during pregnancy and women who were not at high risk using EPDS cut-off score 9 at T1 (Analysis 3 and Analysis 4, respectively). We performed the same logistic regression analyses for the two groups, respectively.
Results
The flow of the number of the participants is shown in Fig 1. A total of 1,775 women gave informed consent. Of them, 1,717 completed the T1 questionnaire and 1,383 completed the T2 questionnaire. Table 1 shows the demographic characteristics of the sample. The data of the women who completed the T2 were used for the bivariate analyses (some data with missing values were excluded in each analysis). Since 250 of them had missing values, the data of 1,133 women were used for the multivariate analyses.
The results of the bivariate analysis are shown in Table 2. Finally, the factors which showed statistical significances (p <0.05) in the bivariate analyses were input into the logistic regression analyses. These factors included the important factors which Robertson and O’Hara mentioned in the previous studies [1, 10].
The results of Analysis 1 are shown in Table 3. The EPDS score (p value (p) <0.01, Odds ratio (OR) = 5.45 [95% confidential interval of the odds ratio (95%CI)] = 3.22–9.22) and primipara (p <0.01, OR = 3.38[95% CI = 2.09–5.45]) were the statistically significant risk factors of PND.
The results of Analysis 2 are shown in Table 4. Significant risk factors include “a perceived lack of family cohesion” (p <0.01, OR = 1.58[95% CI = 1.16–2.15]), primipara (p < 0.01, OR = 3.06[95% CI = 1.94–4.81]), “current physical illness treatment” (p = 0.02, OR = 2.02[95% CI = 1.12–3.64]), and “current psychiatric illness treatment” (p = 0.03, OR = 3.94[95% CI = 1.01–15.39]).
The results of coefficiency statistics for multicollinearity are shown in Table 5. The estimation of the tolerance value and VIF in the models was less than 0.4 and 2.5, respectively, which revealed that there was no multicollinearity in Analysis 1 and Analysis 2 models.
The results of the subanalyses (Analyses 3 and 4) are shown in Table 6. “Divorce” was excluded in Analyses 3 and 4 because of low frequency with the one independent variable. “Frequency of gambling in a month” was also excluded in Analysis 3 because there were no participants who took up gambling in those at high risk of depression at T1. There was no statistical significant predictor, but “leaving or losing one’s job” was a marginal significant predictor in Analysis 3. In analysis 4, “a perceived lack of family cohesion” (AOR 1.55, 95% CI 1.05–2.29, p value 0.03) and “primipara” (AOR 5.52, 95% CI 2.94–10.38, p value<0.01) were shown as significant predictors of PND.
Discussion
Our study has two central findings. First, EPDS can be used as a strong mental health screening tool at 20 weeks gestation to predict PND. Second, there are four important psychosocial risk factors of PND: i.e. a perceived lack of family cohesion, primipara, current physical illness treatment history, and current psychiatric illness treatment history. It is a new finding that a perceived lack of family cohesion is an important psychosocial risk factor of PND. Interestingly, we also found that family cohesion is a more important factor for predicting PND compared to partner support. The other findings are consistent with previous reports.
We show that EPDS is a strong predictor of PND, further supporting the usefulness of antenatal mental health screening. EPDS has been used in the antenatal setting as the Edinburg Depression Scale (EDS) [21, 22]. In Japan, it is not common to use EDS for mental health screening. We propose its use in mental health screening so as to facilitate prompt antenatal interventions in women who are symptomatic or “at risk” for PND.
When considering only the psychosocial aspects, factors such as “a perceived lack of family cohesion”, primipara, “current physical illness treatment”, and “current psychiatric illness treatment” were revealed to be very important to predict PND, according to the interview sheets at obstetric outpatient clinics during the pregnancy period. In terms of the strongest risk factors of PND during pregnancy, Robertson et al. demonstrated factors such as depression, anxiety, stressful life events, low level of social support, and previous history of depression risk factor [10], while O’Hara showed past history of psychopathology and psychological disturbance, poor marital relationship and low social support, and stressful life events. In addition, a low social status showed a small but significant predictive relation to PND [1]. The results of our chi-square analyses showed that all these factors are related to PND in Japanese samples.
Both Analyses 1 and 2 revealed the importance of primipara as a risk factor of PND. The results are consistent with previous studies [23–25]. This is likely due to the fact that women who are pregnant for the first time experience greater psychological stresses compared to those have had gone through childbirth. Furthermore, the tasks and responsibilities of childcare are new to primipara women, and may add to the stress level.
A perceived lack of family cohesion on the women’s part is a new risk factor of PND that has never been reported before. In view of this, professionals in maternal and child health should be more attentive and supportive towards the mother’s feelings for her family during pregnancy. The item “a perceived lack of family cohesion” was added to our questionnaire because we wanted to investigate the importance of the women’s personal relationships when it comes to predicting PND. Family cohesion is defined as shared affection, support, helpfulness, and caring among family members [26–28].
Many studies have demonstrated the importance of emotional and practical support for pregnant mental health [10]. Morikawa et al. [16] showed that a larger number of supportive persons during pregnancy helps protect against postpartum depression, but that satisfaction rating with those supports does not. They used the Japanese version of Social Support Questionnaire 6 [29] (J-SSQ) which had been standardized by Furukawa et al [30]. The questionnaire used in this study did not query participants on the number of supportive persons nor the level of satisfaction with the support. In this study, we classified social support for pregnant women into four categories: emotional support by the partner, emotional support by others besides the partner, practical support by the partner, and practical support by others besides the partner. J-SSQ’s six items can be considered to be related to either practical supports or emotional support, or both. We found that “a perceived lack of family cohesion” was a more important risk factor than emotional and practical support by the partner. This result reflects the unique characteristics of the Japanese society, whereby working hours are generally longer than other countries [31] and many Japanese men find it difficult to have sufficient time to support their partners. While this may typically affect the mental status of mothers, we found that this is not the case in Japan. Japanese women are known for their sense of self-sacrificial love for their family [32]. Therefore, even when there is a lack of support from their partners, they could still feel happy as long as there is a satisfactory level of family cohesion.
We found that mothers who were under physical illness treatment were likely to have greater mental stress, together with the possibility of having psychosomatic symptoms. Perinatal staffs are usually the ones who check for physical problems in pregnant women. Our results suggest that perinatal stuff should also pay attention to the risk factors of postpartum depression. Physical illness treatment has not been addressed as a risk factor of PND in previous studies [1, 9, 10, 33]. Our unique finding may be due to the characteristics of the public health services in Japan. All Japanese citizens are enrolled in a health insurance system which allows them to receive medical services at reasonable prices (70% of medical fees are subsidized). General hospitals, where obstetricians can seamlessly refer pregnant women with physical illnesses to other department doctors, are also easily accessible. However, further studies are still needed to investigate why this factor is significant in Japan. Some mediating factor (i.e. psychosomatic aspects and social and cultural aspects including public health and medical services) may affect this factor.
From our multivariate analysis, current psychiatric illness treatment was identified as an important risk factor of PND, whereas past psychiatric history was not risk factor. However, there is a large number of literature which suggested that both are important risk factors for PND [34, 35]. This additional information can be used by midwives and obstetricians when tending to pregnant women with psychiatric problems.
The results of the subanalyses suggest that “primipara” and “a perceived lack of family cohesion” are important risk factors for those who did not have antenatal depression. In order words, perinatal professionals for postnatal mental care should pay close attention to these factors regardless of whether the pregnant individual had depression or not. There was a discrepancy of AOR between those who were “at high risk” and those who were “not at high risk” of depression during pregnancy in “a perceived lack of family cohesion”. The results suggested that for those without depression, a perceived lack of family cohesion can be a more severe psychological burden compared to depression. On the other hand, the results also suggested that a perceived lack of family cohesion may sometimes be a protective factor against PND. Those who have depression during pregnancy often have family relationship problems (e.g. low level of intimacy with the husband) [36]. The results suggest that, when depression during pregnancy was coupled with family problems, the pregnant woman may learn to be objective rather than being confronted and anguished by the problems—which could be a protective factor against worsening depression. The results of AOR of those who were “not at high risk” of antenatal depression being much larger than that of those who were “at high risk” suggest that primipara is a major antenatal risk factor for PND, especially for those who were not depressive during pregnancy. The results of AOR gap between “at high risk” and “not at high risk” suggest that, if antenatal depression existed, the condition of women with physical illness may worsen compared to those “without depression”. Collaborative care that integrates mental and physical care [37, 38] for those who have physical illness during pregnancy may be important in preventing PND. The results of AOR of “current psychiatric history” of those who were “not at high risk” being much larger than that of those who were “at high risk” (this was also more than 1) suggest that current psychiatric history is a very important antenatal risk factor even for those who are not at high risk of depression during pregnancy. However, the actual mental status of women without depression during pregnancy can be overlooked by perinatal professionals. We suggest that pregnant women with current psychiatric history, even if they were not depressive, should be carefully monitored for having the potential for high risk of PND.
Strengths and limitations
This cohort study was performed in a heavily populated ward (about 900,000 people). The facilities involved in the study allowed a diverse sample, ranging from those who cannot afford perinatal care fees to those who can access expensive, specialized services. Thus, the results in this study can be regarded as strong evidence for perinatal mental health.
There are, however, several limitations in this study. First, we assessed mental health at T1 and T2 only by EPDS. No clinical diagnosis was made using structured or clinicians’ interviews. Thus, we cannot confirm the participants’ mental health with clinical diagnosis. Nonetheless, the sensitivity and specificity of the cut-off score compared to the clinical diagnosis of major depression using the Japanese version of the Schedule for Affective Disorders and Schizophrenia are reliable (75% and 93%, respectively) [17]. Second, this study may not have included a sample that is representative of the entire Japanese population. The Setagaya ward is a residential area in metropolitan Tokyo. Therefore, the participants’ socioeconomic statuses are relatively higher compared to other areas. In addition, many of them were from nuclear families. In Japan, there is a higher prevalence of nuclear families in the major cities compared to the rural regions, where extended families are often important resources for post-delivery care [2, 39]. Third, the questions on psychosocial factors in the questionnaires used in this study were not ones that have been validated (e.g. social supports, family cohesion, experience of child abuse). Fourth, depression during pregnancy and the postnatal period may be accompanied by underlying bipolar disorder. Previous studies reported that underlying bipolarIIdisorder exists in 13% of women with high levels of depressive symptoms in pregnancy [40] and 22% in the postnatal period [41]. This study examined postpartum depression using EPDS only. We did not examine the possibility of bipolar depression or mixed state in the women with depression. Fifth, we used a cut-off score of 9 for EDS for antenatal depression (AND) in this study. Although a cut-off score of 9 has been validated for postnatal depression in a Japanese women sample, there is no validated cut-off score for AND in Japanese women. Several studies used the cut-off score of 10 [42, 43], 12 [44, 45] and 13 [46–48] antenatally and postanatally. Matthey et al. [49] suggested that the validated cut-off score for PND should be 13 or more, and 15 or more for AND in English-speaking women. Considering that the cut-off score for PND is 9 in Japanese women, the cut-off score for AND in Japanese women is likely to be lower than that in English-speaking women. In addition, the purpose of this study was to investigate the risk factor for PND from a population approach viewpoint. We targeted women at high risk of PND and not those with severe depressive state. Thus, we thought it would be more appropriate to use the unvalidated cut-off score of 9 rather than a higher one. Nonetheless, we acknowledged that the use of an unvalidated score was one limitation of this study. Sixth, the attrition rate of this study is not negligible (about one-third of participants who enrolled in this study were not analyzed in the multivariate analyses). As such, the results may not be representative of the general population.
Suggestions for clinical implications and further research
The risk factors of PND revealed in this study (i.e., mental health, primipara, current psychiatric illness treatment, current physical illness treatment, and family cohesion) should be paid attentions by professionals in maternal and child health. In addition, we suggest that EPDS should be used checked during pregnancy to predict postnatal mental health as well as assessing mental health during pregnancy periods. Interestingly, the U.S. Preventive Services Task Force recommends “screening adults for depression (including postnatal, but not pregnant women) when staff-assisted depression care supports are in place to assure accurate diagnosis, effective treatment, and follow-up” [50]. These mental health screenings should be performed where support systems for those who are at risk of mental health problems are available. Many health professionals have come to realize the necessity of integrating mental health screening into routine primary care for pregnant and postpartum women, as well as to follow up this screening with treatment or referral and with follow-up care [9]. In Japan, one of the main reasons why antenatal mental health screening is not common is because support network systems for women with mental health problems have not been developed enough [51]. We suggest that such networks should be established before preforming antenatal mental health screening. Since our results included Japanese culture-based psychosocial aspects, further cross-cultural study about family cohesion and perinatal mental health are needed. The results of multicollinearity tests suggest that these risk factors are demonstrated independent risk factors for PND. In addition, certain combinations of psychosocial factors may be risk factors for PND [52]. Further research on combination psychosocial factors as risk factors for PND is needed. A validation study for the cut-off score for AND in a Japanese sample should also be performed. Furthermore, it would be worthwhile to reassess the antenatal risk factors for PND using validated cut-off score for EDS and to compare the results with those achieved in this study.
Acknowledgments
We would like to thank Dr. Julian Tang of the Department of Education for Clinical Research, National Center for Child Health and Development, for proofreading, editing and writing part of this manuscript.
Ethics approval
Ethics approval was provided by the Institutional Review Board of the National Medical Center for Children and Mothers. Based on the declaration of Helsinki [53], written informed consent to participate in this study was taken from each participant.
Author Contributions
Conceived and designed the experiments: YT T. Kubo RM KT. Performed the experiments: YT KT T. Koizumi HT EO NK. Analyzed the data: YT EI. Wrote the paper: YT. Supported development of this study design from the viewpoints of experts: KY KK MO.
References
- 1. O'hara MW, Swain AM. Rates and risk of postpartum depression-a meta-analysis. International review of psychiatry. 1996;8(1):37–54.
- 2. Yoshida K, Yamashita H, Ueda M, Tashiro N. Postnatal depression in Japanese mothers and the reconsideration of ‘Satogaeri bunben’. Pediatrics International. 2001;43(2):189–93. pmid:11285079
- 3. Dennis C-L, McQueen K. The relationship between infant-feeding outcomes and postpartum depression: a qualitative systematic review. Pediatrics. 2009;123(4):e736–e51. pmid:19336362
- 4. Flykt M, Kanninen K, Sinkkonen J, Punamäki RL. Maternal depression and dyadic interaction: the role of maternal attachment style. Infant and Child Development. 2010;19(5):530–50.
- 5. Cadzow SP, Armstrong KL, Fraser JA. Stressed parents with infants: reassessing physical abuse risk factors. Child Abuse & Neglect. 1999;23(9):845–53.
- 6. Goodman SH, Rouse MH, Connell AM, Broth MR, Hall CM, Heyward D. Maternal depression and child psychopathology: a meta-analytic review. Clinical child and family psychology review. 2011;14(1):1–27. pmid:21052833
- 7. Brand SR, Brennan PA. Impact of antenatal and postpartum maternal mental illness: how are the children? Clinical obstetrics and gynecology. 2009;52(3):441–55. pmid:19661760
- 8. Gress-Smith JL, Luecken LJ, Lemery-Chalfant K, Howe R. Postpartum depression prevalence and impact on infant health, weight, and sleep in low-income and ethnic minority women and infants. Maternal and child health journal. 2012;16(4):887–93. pmid:21559774
- 9. O'Hara MW, McCabe JE. Postpartum depression: current status and future directions. Annual review of clinical psychology. 2013;9:379–407. Epub 2013/02/12. pmid:23394227.
- 10. Robertson E, Grace S, Wallington T, Stewart DE. Antenatal risk factors for postpartum depression: a synthesis of recent literature. General hospital psychiatry. 2004;26(4):289–95. pmid:15234824
- 11. Patel V, Rahman A, Jacob K, Hughes M. Effect of maternal mental health on infant growth in low income countries: new evidence from South Asia. Bmj. 2004;328(7443):820–3. pmid:15070641
- 12. Lee D, Yip A, Leung T, Chung T. Identifying women at risk of postnatal depression: prospective longitudinal study. Hong Kong Med J. 2000;6(4):349–54. pmid:11177155
- 13. Kitamura T, Yoshida K, Okano T, Kinoshita K, Hayashi M, Toyoda N, et al. Multicentre prospective study of perinatal depression in Japan: incidence and correlates of antenatal and postnatal depression. Archives of Women's Mental Health. 2006;9(3):121–30. pmid:16547826
- 14. Miyake Y, Tanaka K, Sasaki S, Hirota Y. Employment, income, and education and risk of postpartum depression: the Osaka Maternal and Child Health Study. Journal of affective disorders. 2011;130(1):133–7.
- 15. Kokubu M, Okano T, Sugiyama T. Postnatal depression, maternal bonding failure, and negative attitudes towards pregnancy: a longitudinal study of pregnant women in Japan. Archives of women's mental health. 2012;15(3):211–6. pmid:22526407
- 16. Morikawa M, Okada T, Ando M, Aleksic B, Kunimoto S, Nakamura Y, et al. Relationship between social support during pregnancy and postpartum depressive state: a prospective cohort study. Scientific Reports. 2015;5:10520. PMC4448522. pmid:26022720
- 17. Okano T, Murata M, Masuchi S, Tamaki R, Nomura J, Miyaoka , et al. Validity and Reliability of the Japanese version of the Edinburgh Postnatal Depression Scale. Archives of Psychiatric Diagnostics and Clinical Evaluation. 1996;7(4):525–33.
- 18. Cox JL, Holden JM, Sagovsky R. Detection of postnatal depression. Development of the 10-item Edinburgh Postnatal Depression Scale. The British journal of psychiatry. 1987;150(6):782–6.
- 19. Murakami M, Iwai H. Definition and Statistics of Working Poor in Japan. Statistics. 2010; (98):13–24.
- 20.
Allison P. Logistic regression using the SAS system: theory and application, 1999. SAS Institute, Cary, NC.167–8.
- 21. Adouard F, Glangeaud-Freudenthal N, Golse B. Validation of the Edinburgh postnatal depression scale (EPDS) in a sample of women with high-risk pregnancies in France. Archives of Women’s Mental Health. 2005;8(2):89–95. pmid:15883653
- 22. Murray D, Cox JL. Screening for depression during pregnancy with the Edinburgh Depression Scale (EDDS). Journal of Reproductive and Infant Psychology. 1990;8(2):99–107.
- 23. Beck CT. A meta-analysis of predictors of postpartum depression. Nursing research. 1996;45(5):297–303. pmid:8831657
- 24. Munk-Olsen T, Jones I, Laursen TM. Birth order and postpartum psychiatric disorders. Bipolar disorders. 2014;16(3):300–7. Epub 2014/03/19. pmid:24636279.
- 25. Dorheim SK, Bjorvatn B, Eberhard-Gran M. Can insomnia in pregnancy predict postpartum depression? A longitudinal, population-based study. PloS one. 2014;9(4):e94674. Epub 2014/04/16. pmid:24732691; PubMed Central PMCID: PMCPMC3986207.
- 26. Barber BK, Buehler C. Family cohesion and enmeshment: Different constructs, different effects. Journal of Marriage and the Family. 1996:433–41.
- 27. Barbarin O. Measuring basic family processes: Validity and reliability of Family Process Scale. Unpublished manuscript. 1984.
- 28.
Moos RH. The social climate scales: An overview: Consulting Psychologists Press; 1980.
- 29. Sarason IG, Sarason BR, Shearin EN, Pierce GR. A brief measure of social support: Practical and theoretical implications. Journal of social and personal relationships. 1987;4(4):497–510.
- 30. Furukawa T, Harai H, Hirai T, Kitamura T, Takahashi K. Social Support Questionnaire among psychiatric patients with various diagnoses and normal controls. Social psychiatry and psychiatric epidemiology. 1999;34(4):216–22. pmid:10365628
- 31. Iwasaki K, Takahashi M, Nakata A. Health problems due to long working hours in Japan: working hours, workers' compensation (Karoshi), and preventive measures. Industrial health. 2006;44(4):537–40. pmid:17085914
- 32.
Lebra TS. Japanese women: Constraint and fulfillment: University of Hawaii Press; 1985.
- 33. Lancaster CA, Gold KJ, Flynn HA, Yoo H, Marcus SM, Davis MM. Risk factors for depressive symptoms during pregnancy: a systematic review. American journal of obstetrics and gynecology. 2010;202(1):5–14. pmid:20096252
- 34. Boyce P. Risk factors for postnatal depression: a review and risk factors in Australian populations. Archives of Women’s Mental Health. 2003;6(2):s43–s50.
- 35. Chee CY, Lee DT, Chong Y, Tan L, Ng T, Fones CS. Confinement and other psychosocial factors in perinatal depression: a transcultural study in Singapore. Journal of affective disorders. 2005;89(1):157–66.
- 36. Kitamura T, Shima S, Sugawara M, Toda M. Clinical and psychosocial correlates of antenatal depression: a review. Psychotherapy and psychosomatics. 1996;65(3):117–23. pmid:8784941
- 37. Gilbody S, Bower P, Fletcher J, Richards D, Sutton AJ. Collaborative care for depression: a cumulative meta-analysis and review of longer-term outcomes. Archives of internal medicine. 2006;166(21):2314–21. pmid:17130383
- 38. Katon W, Von Korff M, Lin E, Walker E, Simon GE, Bush T, et al. Collaborative management to achieve treatment guidelines: impact on depression in primary care. Jama. 1995;273(13):1026–31. pmid:7897786
- 39.
Ministry of Health LaW. Attitudes for delivery and child care. Year Book of Japanese Ministry of Health, Labour and Welfare. 2013.
- 40. Lydsdottir LB, Howard LM, Olafsdottir H, Thome M, Tyrfingsson P, Sigurdsson JF. The mental health characteristics of pregnant women with depressive symptoms identified by the Edinburgh Postnatal Depression Scale. The Journal of clinical psychiatry. 2014;75(4):393–8. pmid:24569071
- 41. Wisner KL, Sit DK, McShea MC, Rizzo DM, Zoretich RA, Hughes CL, et al. Onset timing, thoughts of self-harm, and diagnoses in postpartum women with screen-positive depression findings. JAMA psychiatry. 2013;70(5):490–8. pmid:23487258
- 42. Barnett B, Glossop P, Matthey S, Stewart H. Screening in the context of integrated perinatal care. Screening for perinatal depression. 2005:68–82.
- 43. Morse CA, Buist A, Durkin S. First-time parenthood: influences on pre-and postnatal adjustment in fathers and mothers. Journal of Psychosomatic Obstetrics & Gynecology. 2000;21(2):109–20.
- 44. Condon JT, Boyce P, Corkindale CJ. The first-time fathers study: A prospective study of the mental health and wellbeing of men during the transition to parenthood. Australian and New Zealand Journal of Psychiatry. 2004;38(1–2):56–64. pmid:14731195
- 45. de Tychey C, Spitz E, Briançon S, Lighezzolo J, Girvan F, Rosati A, et al. Pre-and postnatal depression and coping: a comparative approach. Journal of Affective Disorders. 2005;85(3):323–6. pmid:15780702
- 46. Evans J, Heron J, Francomb H, Oke S, Golding J. Cohort study of depressed mood during pregnancy and after childbirth. Bmj. 2001;323(7307):257–60. pmid:11485953
- 47. Heron J, O'Connor TG, Evans J, Golding J, Glover V, Team AS. The course of anxiety and depression through pregnancy and the postpartum in a community sample. Journal of affective disorders. 2004;80(1):65–73. pmid:15094259
- 48. Leung SS, Martinson IM, Arthur D. Postpartum depression and related psychosocial variables in Hong Kong Chinese women: findings from a prospective study. Research in nursing & health. 2005;28(1):27–38.
- 49. Matthey S, Henshaw C, Elliott S, Barnett B. Variability in use of cut-off scores and formats on the Edinburgh Postnatal Depression Scale—implications for clinical and research practice. Archives of women's mental health. 2006;9(6):309–15. pmid:17013761
- 50. Force UPST. Screening for depression in adults: US preventive services task force recommendation statement. Ann Intern Med. 2009;151(11):784–92. pmid:19949144
- 51.
Tachibana Y. Research report on the promotion for networks cooperating with hospitals, public health, and welfare to support women with depression in the perinatal periods. the Health and Labor Science Research Grant on Psychiatry and Neurological Disease and Mental Health, provided by the Ministry of Health, Labor and Welfare of Japan. 2013: 1–6 (in Japanese).
- 52. Rubertsson C, Wickberg B, Gustavsson P, Rådestad I. Depressive symptoms in early pregnancy, two months and one year postpartum-prevalence and psychosocial risk factors in a national Swedish sample. Archives of Women’s Mental Health. 2005;8(2):97–104. pmid:15883652
- 53. Rickham P. Human experimentation. Code of Ethics of the World Medical Association. Declaration of Helsinki. British Medical Journal. 1964;2(5402):177. pmid:14150898