article critique- Postpartum depression

ORIGINAL ARTICLE

Violence as a risk factor for postpartum depression in mothers: a meta-analysis

Qian Wu & Hong-Lin Chen & Xu-Juan Xu

Received: 9 September 2011 /Accepted: 11 December 2011 /Published online: 1 March 2012 # Springer-Verlag 2012

Abstract The objective of this paper is to examine the association between violence and postpartum depression (PPD). The data sources of this study are: Web of Science, PubMed, Elsevier, Springer Link were exam- ined from their start date through July1, 2011. “Vio- lence”, “domestic violence”, “physical violence”, “sexual violence”, “domestic violence”, “postpartum de- pression”, “postnatal depression”, and “puerperal depres- sion” were some of the terms included in the purview of MeSH terms. Relevant studies from reference lists were also scanned. Studies examining the association between violence and postpartum depression have been included. A total of 679 studies were included in this screening. Essential information of these included stud- ies was independently extracted by two raters. Newcas- tle–Ottawa scale was used to assess the clinical data of these research studies. Random-effects model was cho- sen in this meta-analysis for maintaining significant heterogeneity. Publication bias was evaluated with the help of a funnel plot. Six studies involving 3,950 par- ticipants were included in this clinical study. Violence was one of the factors responsible for PPD [OR03.47; 95% confidence interval (CI; 2.13–5.64)]. Significant heterogenity was found in this meta-analysis (P<

0.00001; I2079%)and publication bias was detected through a funnel plot. A sensitivity analysis of 3.00, 95%CI (2.44–3.68), p<0.00001 indicated that our find- ings were robust and reliable. Our meta-analysis indi- cated a positive correlation between violence and PPD. In women of reproductive age, PPD induced through violence can be prevented through early identification. In addition, due to heterogeneity and wide CIs in this meta-analysis, further research is evidently required.

Keywords Violence . Postpartum depression .

Meta-analysis

Introduction

Postpartum depression (PPD) is a common psychiatric illness in many women of diverse cultures. It has been identified as a major public health problem. The preva- lence of PPD varies in different countries. Pregnancy risk monitoring system has been used for evaluating postpartum depression in women. In the period extend- ing from 2004–2005, 143 clinical studies from 40 countries have reported that PPD has increased from 0% to 66% (Cantilino et al. 2010). Approximately 6.5–13% of women experience this condition in the first year (Boyd et al. 2011).

PPD can hamper a mothers’ ability to cope with life events, including parenting tasks. Mothers experience emotions of mood swings, anxiety, guilt, and low self- esteem. Some women with postpartum depression may even experience suicidal tendencies. PPD has detrimental effects on not only the mother, but also the child. Re- search studies have suggested that depressive symptom- atology in the postpartum period increases the

Qian Wu and Hong-Lin Chen contributed equally to the study.

Q. Wu : H.-L. Chen Nursing school of Nantong University, Nantong, People’s Republic of China

X.-J. Xu (*) Affiliated Hospital of Nantong University, Xi Si Road 20#, Nantong City, Jiangsu Province 226001, People’s Republic of China e-mail: [email protected]

Arch Womens Ment Health (2012) 15:107–114 DOI 10.1007/s00737-011-0248-9

susceptibility to developing negative emotional outcomes (Dennis and McQueen 2009). In developing countries, poor maternal mental health has been associated with infant malnutrition and reduced physical health (Rahman et al. 2002). Depressed mothers are also less sensitive to their infant’s needs, and are more likely to express nega- tive behaviors. This has been associated with lower cog- nitive functioning and adverse emotional development in children (Abbaszadeh et al. 2011).

Many factors are responsible for inducing PPD. According to research studies, some of the significant factors causing depression include a family history of depression, stressful situation during pregnancy or early puberty, anxiety during pregnancy, and low social support (Robertson et al. 2004). In recent times, some findings (Henwood 2000; Jasinski 2004; Torres S et al. 2000) indicate that not only somatic complaints but also anxiety

and post-traumatic stress disorder (PTSD) could be caused through violence. Women with PTSD feel anxious and depressed. Furthermore, violence against women (VAM) is considered as a chronic stressor (Latendresse 2009). When it exceeds an individual’s capacity to re- spond, mental health problems such as depression may also occur (Beydoun et al. 2010). Some articles have reported that violence against women (VAW) increases the incidence of PPD (Urquia et al. 2011). Therefore, VAM may be another trigger of PPD.

According to the United Nations definition, the term VAM implies any act of gender-based violence that results in, or is likely to result in, physical, sexual, or psychological harm or suffering to women. This also includes threatening, coercion, and arbitrary deprivation of liberty in public and private life (Elsberg and Heise 2005). According to a WHO multi-country study, the reported lifetime prevalence of physical or sexual violence varies from 15% to 71% in ten countries (Garcia-Moreno et al. 2006). VAM has been an important criteria of global public health problem, and women of reproductive age are at the heart of this issue (Campbell 2002).

Numerous research studies have assessed the relation between violence and PPD. However, due to indistinct forms of violence, the findings of these studies are not quite clear. Records and Cohen et al. have failed to indicate that physical or sexual abuse is a significant variable of PPD (Records and Rice 2005; Cohen et al. 2002). On the other hand, Buist and Rice et al. have suggested that physical or sexual abuse is one of the pre-disposing factors for PPD (Buist and Janson 2001; Rice et al. 2001). At present, considering the hazards of VAM and PPD, there is a paucity of meta-analysis in the previous studies examining the effect of exposure to violence on PPD. We conducted a meta- analysis to examine the association between all forms of violence and PPD.

Method

Search strategy

A literature search was conducted to ascertain the association between violence and PPD. Searches were performed in Web of Science, PubMed, Elsevier, and SpringerLink. We per- formed the following literature search: 1. ((“violence”[Mesh]) OR “domestic Violence”[Mesh]) AND “depression, postpar- tum/etiology” [Mesh] 2. (((physical violence) OR sexual vio- lence) OR domestic violence)AND (((postpartum depression) OR postnatal depression) OR puerperal depression). Each database was searched thoroughly from its start date through to July 1, 2011. All published citations were cross-referenced

Articles identified through literature search , n=676

PubMed n=88; Web of Science n=68;

Science Direct n=26; Springerlink n=494

Reference listing of relevant

articles identified 13

additional articles

689 abstracts and titles were screened using the

inclusion and exclusion criteria

Excluded n=666

Review articles n=42

Uncorrelated articles n=326

No full-text n=277

Duplicate studies n=21

23 full text articles screened

Excluded 1 7 for no sufficient

data

6 articles included for data extraction and analysis

Fig. 1 Flow chart of studies selection

108 Q. Wu et al.

T a b le

1 C h ar ac te ri st ic s o f in cl u d ed

st u d ie s

A u th o rs

Y ea r

S tu d y

lo ca ti o n

D es ig n

S am

p le sa

A g e

T y p e o f

v io le n ce

H o w

v io le n ce

as se ss ed

W h en

P P D

as se ss ed

H o w

P P D

as se ss ed

R es u lt

C er ta in

et al .

2 0 0 8

U S A

C ro ss -s ec ti o n al

1 ,5 1 9 (1 ,6 3 3 )

≥ 1 8

D o m es ti c v io le n ce

A A S b

4 5 d ay s

p o st p ar tu m

E P D S c

O R 4 .2 1 9 5 % C I

(2 .1 9 – 8 .0 9 )

D o la ti an

et al .

2 0 1 0

Ir an

P ro sp ec ti v e

co h o rt

2 4 0 (2 5 1 )

1 5 – 4 0

A ll ty p es

A v al id at ed

q u es ti o n n ai re

2 – 6 w ee k s

p o st p ar tu m

E P D S c

O R 5 .8 8 9 5 % C I

(2 .1 – 5 .1 )

L eu n g

et al .

2 0 0 2

C h in a

P ro sp ec ti v e

co h o rt

8 3 8 (8 7 5 )

n o d at a

D o m es ti c v io le n ce

A A S b

2 – 3 d ay

an d 6 w ee k s

P o st p ar tu m ; 1 – 2

d ay s af te r d is ch ar g e

D S S E P D S c

O R 1 .7 9

L u d er m ir

et al .

2 0 1 0

B ra zi l

P ro sp ec ti v e

co h o rt

1 ,0 4 5 (1 ,1 3 3 )

≥ 1 8

P sy ch o lo g ic al

/p h y si ca l/ se x u al

v io le n ce

A v al id at ed

q u es ti o n n ai re

3 – 6 m o n th s p o st p ar tu m

E P D S c

O R 2 .1 2

9 5 % C I 2 3 .2 – 2 8 .6

S av ar im

u th u

et al .

2 0 1 0

In d ia

Q u al it at iv e

m et h o d o lo g y

1 2 5 (1 3 7 )

n o d at a

P h y si ca l v io le n ce

S E M Ib

2 – 1 0 w ee k s

p o st p ar tu m

E P D S c

O R 3 .8 8 , 9 5 % C I

(1 .3 6 – 11 .0 );

O R 3 .3 3 , 9 5 % C I

(1 .0 9 – 9 .6 1 )

A li 1 et

al .

2 0 0 9

P ak is ta n

Q u as i- ex p er im

en ta l

st u d y

2 6 7 (4 2 0 )

n o d at a

D o m es ti c v io le n ce

H E /F R Q b

1 , 2 , 6 , an d 1 2

m o n th s p o st p ar tu m

A K U A D S c

O R 2 .7 7 (1 .2 2 – 6 .2 8 )

aO R 2 .8 2 (1 .1 5 –

6 .8 9 )

a S am

p le s in d ic at e th e n u m b er

o f p ar ti ci p an ts w h o w er e el ig ib le

fo r th e st u d ie s, th e n u m b er

in b ra ck et s in d ic at e th e n u m b er

o f w o m en

w h o co m p le te d th e w h o le

st u d y

b A A S A b u se

A ss es sm

en t sy st em

; S E M I sh o rt E x p la n at o ry

M o d el

In te rv ie w ; H E /F R Q

H o m e en v ir o n m en t/ F am

il y re la ti o n sh ip

q u es ti o n n ai re

c E P D S E d in b u rg h p o st p ar tu m

D ep re ss io n S ca le ; A K U A D S A g a K h an

U n iv er si ty

A n x ie ty

an d D ep re ss io n S ca le ; D S S D ai ly

S co ri n g S y st em

Violence as a risk factor for postpartum depression in mothers 109

from other relevant studies. The clinical study was conducted in English

Study selection

The present clinical study addresses the specific issue of whether violence is associated with an increased risk of

PPD. Research studies focusing on the relation between PPD and violence have been included. Quite a few research studies were conducted to include a standardized assessment of depression [e.g., Edinburgh Postnatal Depression Scale (EPDS)] and violence [e.g., Abuse Assessment Screen (AAS), Family relationship questionnaire (FRQ)]. Adequate statistics are required to submit a report on meta-analysis

Table 2 Assessment of methodological quality of prospective cohort studies

Reference Selection Comparabilitye Outcome

Representativeness of the exposed cohorta

selection of the non- exposed cohortb

Ascertainment of exposurec

Incident diseased

Assessment of outcomef

Length of follow-upg

Adequacy of follow-uph

Ludermir ★ ★ ★ ★★ ★ ★

Dolatian ★ ★ ★ ★★ ★ ★ ★

Leung ★ ★ ★ ★★ ★ ★ ★

Ali ★ ★ ★ ★ ★

a A, truly representative of the average patient with postpartum depression (PPD); B, somewhat representative of the average patient with PPD; C, selected group; D, no description of the derivation of the cohort b A, drawn from the same community as the exposed cohort; B, drawn from a different source; C, no description of the derivation of the non- exposed cohort c A, secure record; B, structured interview; C, written self-report; D, no description d For demonstration that the outcome of interest was not present at start of study: A, yes; B, no e For comparability of cohorts on the basis of the design or analysis: A, study controls for the most important factor; B, study controls for any additional factor f A, independent blind assessment; B, record linkage; C, self-report; D, no description g For determination of whether follow-up was long enough for outcomes to occur: A, yes; B, no h A, complete follow-up and all subjects accounted for; B, subjects lost to follow-up was unlikely to introduce bias, because a small number were lost (i.e. >90% were available for follow-up) or a description was provided of those lost; C, follow-up rate of <90% (select an adequate percentage) and no description of those lost; D, no statement

Table 3 Assessment of methodological quality of cross-sectional studies

Reference Selection Comparabilitye Exposure

Case definitiona

Representativeness of the casesb

Selection of controlsc

Definition of controlsd

Ascertainment of exposuref

Ascertainment for cases and controlsg

Non- response rate h

Savarimuthu ★ ★ ★ ★ ★★ ★ ★

Certain ★ ★ ★ ★ ★ ★ ★

a A, case definition is adequate with independent validation; B, case definition is adequate with record linkage or based on self-reports; C, no description. b A, consecutive or obviously representative series of cases; B, potential for selection biases or not stated. c A, community controls; B, hospital controls; C, no description. d A, no history of disease; B, no description of source e Comparability of cases and controls on the basis of the design or analysis: A, study controls for the most important factor; B, study controls for any additional factor. f A, secure record; B, structured interview where blind to case/control status; C, interview not blinded to case/control status; D, written self-report or medical record only; E, no description. g Some method of ascertainment for cases and controls: A, yes; B, no. h A, same rate for both groups; B, non respondents described; C, rate different and no designation

110 Q. Wu et al.

procedures. Articles were excluded if: (1) provided data originated from case report, reviews, or meta-analysis; (2) full text could not be obtained through many ways; (3) sufficient data was not reported (Fig. 1).

Assessment of study quality

Although quality assessment can be reliably conducted in meta-analyses of experimental studies, its use in operation research is quite controversial, with no clear consensus on rating methods or appropriate use in analysis (Paulson et al. 2010). In this meta-analysis, Newcastle–Ottawa Scale (NOS) was used for assessing the quality of non-randomized studies. On the other hand, meta-analysis was used to assess the quality of articles. With this tool, a study is judged on three broad perspectives: the selection of study groups, the compa- rability of groups, and ascertainment of either the exposure or outcome of interest for case–control or cohort studies .This tool identifies ‘high’ quality choices with a “star”. A maxi- mum of one “star” for each item is assigned within the “Selection” and “Exposure/Outcome” categories. On the other

hand, a maximum of two “stars” is assigned for “Compara- bility”. Two reviewers have independently assigned a quality rating to each trial. Results were compared and differences were discussed until an agreement was reached.

Data extraction

Two reviewers have independently extracted data and recorded information of authors, publication year, study de- sign, participants (number and characteristics), location, out- comes assessment, time of outcome measurement, and results. Any disagreement between reviewers was resolved through discussion and referred to a third party. On completion of data extraction, no major discrepancies between the reviewers were identified in the data.

Data synthesis

We conducted a meta-analysis of the relation between vio- lence and PPD. Review Manager 5.0 was used to conduct this meta-analysis. Statistical tests determining heterogeneity is very popular in meta-analysis. They are commonly used as methods for combining studies conforming to the consistency or inconsistency of findings (Higgins et al. 2002; Petitti 2001). Statistical heterogeneity was assessed using I2 statistic, where- in values of 50% or more indicated a substantial level of heterogeneity (Higgins et al. 2003). I2 statistic describes the percentage of variation across studies that are due to hetero- geneity rather than chance (Higgins and Thompson 2002; Higgins et al. 2003). A random-effects meta-analysis was used when it indicated a high level among the trials included in the analysis. We have chosen “Inverse Variance” method to cal- culate the odds ratio, the 95% confidence interval, and the p value of each study. Sensitivity analysis has been applied to assess the robustness of results to the exclusion of trials. It is the study of how the variation (uncertainty) in the output of a

Fig. 2 Meta-analysis of the effect of violence on puerperal depression. Random-effects model was used to estimate the overall effects. OR odds ratio; 95%CI the 95% confidence interval; Chi2 the value of a χ2

test, a p value of less than 0.05 is often used to indicate heterogeneity; I2 the value of quantitative analysis of heterogeneity

Fig. 3 Funnel plot used to assess the publication bias

Violence as a risk factor for postpartum depression in mothers 111

statistical model could be attributed to different variations in the inputs of the model (Saltelli et al. 2009). Publication bias was estimated by preparing a funnel plot. Publication bias may lead to asymmetrical funnel plots.

Results

Trail flow and characteristics of included studies

Figure 1 summarizes the clinical research studies included in our meta-analysis. The combined searches identified 676 potentially relevant clinical studies, while reference listing of relevant articles identified 13 additional articles. Most of these research studies (n0666) were excluded because they were not consistent with our meta-analysis. These research studies focused on other topics or just focused on PPD or violence, reviews, and duplicate studies. After further re- view, six articles that met the inclusion criteria were includ- ed in this review (Ali et al. 2009; Certain et al. 2008; Leung et al. 2002; Ludermir et al. 2010; Savarimuthu et al. 2010; Dolatian et al. 2010).

The six trials included in the meta-analysis incorporated 3,950 women. These articles were published in the period extending from 2002 and 2010. Table 1 shows a summa- rized description of these included studies. The table pro- vided different parameters of information for each study: sample size, publication dates, study design, a description of the participants, when PPD assessed, how PPD assessed, method for collecting data about violence, and results of these studies.

Methodological quality

Of the six trials included for meta-analysis, one is a quasi-experimental study (Ali et al. 2009), the quality of which cannot be assessed by Newcastle–Ottawa Scale, because this tool can only be used in non-randomized studies for quality evaluation. Leung et al. (2002) and Ludermir et al. (2010) are prospective cohort studies and the remaining studies (Certain et al. 2008; Dolatian et al. 2010; Savarimuthu et al. 2010) are cross-sectional sur- veys. Their quality evaluation is performed through a “star system” of Newcastle–Ottawa Scale, the details of which can be seen in Tables 2 and 3.

Meta-analyses of selected studies

While comparing the prevalence of PPD between violence group and non-violence group, this meta-analysis showed a significantly increased risk of PPD in violence group. 576 patients were included in the violence group, while 3,374 patients were included in the non-violence group. By

looking at the forest plot, we found a pooled OR of 3.47 [95%CI (2.13–5.64)] for the overall. The heterogeneity in the meta-analysis was statistically significant and large in magnitude (P<0.00001; I2079%; Fig. 2). Publication bias was assessed through a funnel plot, which is primarily used as a visual aid for detecting bias. In Fig. 3, an asymmetrical funnel plot indicates that there is a publication bias in this review.

The approach to sensitivity analysis of this review is to change the model and observe its behavior. A fixed-effects meta-analysis was used with OR 3.00, 95%CI (2.44–3.68), p<0.00001(Fig. 4). No obvious difference between a random-effects model and a fixed-effects model was ob- served. So, the result of this study is stable and reliable.

Discussion

PPD is one of the most common complications in women of childbearing age. Approximately 10–15% of new mothers experience this condition (Munk-Olsen et al. 2006). PPD can cause psychiatric morbidity among mothers, triggering adverse effects on their babies (Hayes et al. 2010). Therefore, PPD is a matter of great concern. Some articles have indicated that VAM may be a predictor of PPD (Valentine et al. 2011). VAW is the most common form of gender-based violence in the world. It is regarded as a chronic stressor, because it has a detrimental effect on the mental health of these women, espe- cially those in reproductive age. We have conducted this meta- analysis to evaluate the role of violence on PPD.

The meta-analysis of quite a few research studies has highlighted a positive association between violence and PPD (OR03.47; 95%CI: 2.13–5.64). This is consistent with the results of six studies included in this meta-analysis. However, most women suffering from violence and PPD do not seek the services of medical practitioners. In fact, most of them do not get any assistance from their family and friends (Sen 2004). Many people are of the view that report- ing domestic violence is an embarrassing incidence (Kim et

Fig. 4 Sensitivity analyses for the effect of violence on puerperal depression. Fixed-effects model was used to estimate the overall effects. OR odds ratio; 95%CI the 95% confidence interval; Chi2 the value of a χ2 test, a p value of less than 0.05 is often used to indicate heterogeneity; I2 the value of quantitative analysis of heterogeneity

112 Q. Wu et al.

al. 2011; Zarif 2011). Although those puerperants can re- ceive various treatment options from medical practitioners, they tend to conceal the psychological problems associated with acts of violence. Weight loss and sleep problems are some of the typical symptoms experienced by women suf- fering from PPD (Saurel-Cubizolles et al. 2000), but their caregivers frequently sidelined PPD, whereas the diagnosis were often skipped in obstetrics. This ensures that the patients failed to receive support services and appropriate timely treatment, leading to serious psychological and social consequences in some cases. Sometimes, it can lead to infanticide, maternal death, and suicide (Almond 2009). A range of services should be funded, including hotline for counseling services. We need to set up a refuge for abused women. Psychological guidance, legal, and medical services should be provided to protect these women from violence. In high-risk populations, a screening tool, which is simple, rapid, and accurate, is required for early identification of PPD. Early identification and appropriate psychotherapeutic interventions can significantly reduce the adverse effects in mother, infant, and family (Burton et al. 2011). The symp- toms of PPD can also be minimized with the help of a supportive partner and support system (Gremigni et al. 2011). American Academy of Pediatrics states that the pri- mary care clinician plays a pivotal role in surveillance, screening, and working with families to improve the expec- tant mothers’ mental health problems. Therefore, we need quite a few well-trained professionals focusing in this area.

The heterogeneity of six trails was quite substantial. Demographic characters have accounted for considerable heterogeneity. Some of the factors contributing to PPD include age (less than 20 or more than 30 years), low education, economic difficulties, and unhappy marriage (Ludermir et al. 2010). On the other hand, according to the clinical study conducted by Savarimuthu et al. 2010 that age and level of education were irrelevant to PPD. Therefore, further research is required to examine their correlation. PPD can begin at any time between delivery and 6 months after delivering. Sometimes, it can last up to 1 year. It peaks within 3–6 month after parturition. However, the measuring tool for PPD and the timing of measurement in the six studies differs, bringing out heterogeneity. The heterogene- ity is also attributed to the difference in the evaluation method in terms of history of violence, design of study, and study location. Initially, to prevent heterogeneity in our study, we conducted subgroup analyses. But, research studies did not provide sufficient data and the number of studies in each category was finite. Therefore, subgroup analysis cannot be available.

The asymmetrical funnel plot may be indicative of pub- lication bias. Publication bias is a tendency in which signif- icant results are produced while negative or near neutral ones are seldom published. It can also be caused by chance

element and some other reasons. Sensitivity analysis sug- gests that our findings are quite reliable. Several parameters of violence have been included in this review. These clinical research studies have included three forms of violence, namely, domestic violence, physical violence, and sexual violence. This fails to establish how a specific act of vio- lence influences PPD. Variable tools measuring PPD were used in different time quanta, thereby limiting the interpre- tation of time-specific prevalence in PPD. Furthermore, studies included in this article are consistent with those with sufficient data. On the other hand, a few null studies have been excluded owing to insufficient data. This leads to a potential bias in our findings.

In summary, we have found some evidence to reveal the link between violence and PPD. We need to devise strategies and measures which are capable of protecting women from violence. Future research should focus on methods which could easily identify and prevent PPD in high-risk popula- tions, especially those suffered from violence. We hope that these measures decrease the occurrence of postpartum depres- sion and ultimately improve the health of women.

Conflict of interest None.

References

Abbaszadeh A, Kermani FP, Safizadeh H (2011) Violence during pregnancy and postpartum depression. Pak J Med Sci 27:177–181

Ali NS, Ali BS, Azam IS (2009) Post partum anxiety and depression in peri-urban communities of Karachi, Pakistan: a quasi-experimental study. BMC Publ Health 9:384

Almond P (2009) Postnatal depression: a global public health perspec- tive. Perspect Public Health 129(5):221–227

Beydoun HA, Al-Sahab B, Beydoun MA, Tamim H (2010) Intimate partner violence as a risk factor for postpartum depression among Canadian women in the Maternity Experience Survey. Ann Epi- demiol 20(8):575–583

Boyd RC, Mogul M, Newman D, Coyne JC (2011) Screening and referral for postpartum depression among low-income women: a qualitative perspective from community health workers. Depress Res Treat 2011:320605

Buist A, Janson H (2001) Childhood sexual abuse, parenting and postpartum depression—A 3-year follow-up study. Child Abuse & Neglect 25(7):909–921

Burton A, Patel S, Kaminsky L, Rosario GD, Young R, Fitzsimmons A, Canterino JC (2011) Depression in pregnancy: time of screen- ing and access to psychiatric care. J Matern Fetal Neonatal Med 24(11):1321–1324

Campbell JC (2002) Health consequences of intimate partner violence. Lancet 359(9314):1331–1336

Cantilino A, Zambaldi FC, Lcena T (2010) Postpartum depression on Recife-Brazil: prevalence and association with bio-socio- demographic factors. J Bras Psiquiatr 59:1–9

Certain HE, Mueller M, Jagodzinski T, Fleming M (2008) Domestic abuse during the previous year in a sample of postpartum women. J Obstet Gynecol Neonatal Nurs 37(1):35–41

Violence as a risk factor for postpartum depression in mothers 113

Cohen MM, Schei B, Ansara D, Gallop R, Stuckless N, Stewart DE (2002) A history of personal violence and postpartum depression: is there a link? Archives of Women’s. Mental Health 4(3):83

Dennis CL, McQueen K (2009) The relationship between infant- feeding outcomes and postpartum depression: a qualitative sys- tematic review. Pediatrics 123:736–751

Dolatian M, Hesami K, Shams J, Majd HA (2010) Relationship be- tween violence during pregnancy and postpartum depression. Iranian Red Crescent Med J 12:377–383

Elsberg M, Heise L (2005) Researching violence against women: a practical guide for researches and activists. World Health Organi- zation, PATH, Washington

Garcia-Moreno C, Jansen HA, Ellsberg M, Heise L, Watts CH, WHO Multi-country Study on women’s Health and Domestic Violence against Women Study Team (2006) Prevalence of intimate partner violence:findings from the WHO multi-country study on women’s health and domestic violence

Gremigni P, Mariani L, Marracino V, Tranquilli AL, Turi A (2011) Partner support and postpartum depressive symptoms. J Psycho- som Obstet Gynaecol 32(3):135–140

Hayes DK, Ta VM, Hurwitz EL, Mitchell-Box KM, Fuddy LJ (2010) Disparities in self-reported postpartum depression among Asian, Hawaiian, and Pacific Islander Women in Hawaii: Pregnancy Risk Assessment Monitoring System (PRAMS), 2004–2007. Matern Child Health J 14(5):765–773

Henwood M (2000) Domestic violence: a resource manual for health care professionals. Department of Health, London

Higgins JP, Thompson SG (2002) Quantifying heterogeneity in a meta- analysis. Stat Med 21(11):1539–1558

Higgins J, Thompson S, Deeks J, Altman D (2002) Statistical hetero- geneity in systematic reviews of clinical trials: a critical appraisal of guidelines and practice. J Health Serv Res Policy 7(1):51–61

Higgins JP, Thompson SG, Deeks JJ, Altman DG (2003) Measuring inconsistency in meta-analyses. BMJ 327(7414):557–560

Jasinski JL (2004) Pregnancy and domestic violence: a review of the literature. Trauma Violence Abuse 5(1):47–64

Kim S, Thibodeau R, Jorgensen RS (2011) Shame, guilt, and depressive symptoms: a meta-analytic review. Psychol Bull 137(1):68–96

Latendresse G (2009) The interaction between chronic stress and pregnancy: preterm birth from a biobehavioral perspective. J Midwifery Womens Health 54(1):8–17

Leung WC, Kung F, Lam J, Leung TW, Ho PC (2002) Domestic violence and postnatal depression in a Chinese community. Int J Gynaecol Obstet 79(2):159–166

Ludermir AB, Lewis G, Valongueiro SA, de Araujo TV, Araya R (2010) Violence against women by their intimate partner during

pregnancy and postnatal depression: a prospective cohort study. Lancet 376(9744):903–910

Munk-Olsen T, Laursen TM, Pedersen CB, Mors O, Mortensen PB (2006) New parents and mental disorders: a population-based register study. JAMA 296:2582–2589

Paulson JF, Bazemore SD (2010) Prenatal and Postpartum Depression in Fathers and Its Association With Maternal Depression A Meta- analysis. Clinician’s Corner 303(19):1961–1969

Petitti DB (2001) Approaches to heterogeneity in meta-analysis. Stat Med 20(23):3625–3633

Rahman A, Harrington R, Bunn J (2002) Can maternal depression increase infant risk of illness and growth impairment in develop- ing countries? Child Care Health Dev 28(1):51–56

Records K, Rice MJ (2005) A comparative study of postpartum de- pression in abused and non-abused women. Arch Psychiatr Nurs 19(6):281–290

Rice M, Records K, Williams M (2001) Postpartum depression: iden- tification treatment and prevention. Clin Let Nurse Pract 5:1–4

Robertson E, Grace S, Wallington T, Stewart DE (2004) Antenatal risk factors for postpartum depression: a synthesis of recent literature. Gen Hosp Psychiatry 26(4):289–295

Saltelli A, Ratto M, Andres T, Campolongo F, Cariboni J, Gatelli D, Saisana M, Tarantola S (2009) Global Sensitivity Analysis. Biometrics 65(4):1311–1312

Saurel-Cubizolles MJ, Romito P, Lelong N, Ancel PY (2000) Women’s health after childbirth: a longitudinal study in France and Italy. BJOG 107(10):1202–1209

Savarimuthu RJ, Ezhilarasu P, Charles H, Antonisamy B, Kurian S, Jacob KS (2010) Post-partum depression in the community: a qualitative study from rural South India. Int J Soc Psychiatry 56(1):94–102

Sen B (2004) Adolescent propensity for depressed mood and help seeking: race and gender differences. J Ment Health Policy Econ 7(3):133–145

Torres S, Campbell J, Campbell DW (2000) Abuse during and before pregnancy: Prevalence and Cultural Correlates. Violence Vict 15 (3):303–321

Urquia ML, O’Campo PJ, Heaman MI, Janssen PA, Thiessen KR (2011) Experiences of violence before and during pregnancy and adverse pregnancy outcomes: an analysis of the Canadian Maternity Experiences Survey. BMC Preg Childbirth 11:42

Valentine JM, Rodriguez MA, Lapeyrouse LM, Zhang M (2011) Recent intimate partner violence as a prenatal predictor of mater- nal depression in the first year postpartum among Latinas. Arch Womens Ment Health 14(2):135–143

Zarif M (2011) Feeling shame: insights on intimate partner violence. J Christ Nurs 28(1):40–45

114 Q. Wu et al.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.