Pathways Linking Intimate Partner Violence and Postpartum Depression Among Jewish and Arab Women in Israel

Abstract

Intimate partner violence (IPV) is a major risk factor for postpartum depression (PPD), with 9% to 28% of PPD cases reporting IPV at some point in their lives. Yet little is known about how these phenomena are associated. We asked, “What direct and indirect pathways link IPV to PPD in women belonging to different ethnic–national groups in Israel?” We recruited a stratified sample of Jewish and Arab women, 18 to 48 years old and 6 months postpartum, during their visits to maternal and child health clinics. We computed path analyses to identify both direct and indirect predictors linking IPV frequency and PPD in a stratified sample of Jewish (n = 807) and Arab (n = 248) women. The overall rate of PPD was estimated at 10.3%, whereas the rate of IPV for the total sample was 36%. We identified a direct link between IPV and PPD. IPV also appeared to have an equivalent, indirect effect on PPD via greater chronic stress and reduced social support. IPV was greater and social support was lower for Arab women, who also reported higher PPD, independent of sociodemographic differences between ethnic groups (i.e., education, occupation). Of note, an unplanned pregnancy appeared to increase the risk of both IPV and PPD. Our findings suggest that complex pathways link IPV to PPD and that indirect effects of IPV are equivalent to its direct effects on postpartum women. These findings contribute to a growing international body of research showing the significant effects of IPV on health and well-being. The factors we identified as directly and indirectly associated with PPD might inform interventions to identify and treat PPD.

Keywords

intimate partner violence postpartum depression path analysis social support Arab and Jewish women in Israel

Introduction

Roughly 15% of women will experience postpartum depression (PPD) in the first year after childbirth (Faisal-Cury et al., 2013). Among minority women, rates of PPD are higher (Ceballos et al., 2016; Haque et al., 2015; Liu & Tronick, 2013). PPD among mothers usually begins within the first 6 weeks after childbirth. In most cases, the experience warrants clinical intervention, to which PPD tends to respond (Robertson et al., 2004). In contrast to other forms of depression, PPD is believed to be largely caused by hormonal and other physiological changes associated with pregnancy and childbirth (Surkan et al., 2006). However, PPD is also associated with various psychological, socioeconomic, and cultural factors (Deng et al., 2014; Haque et al., 2015).

In Israel, PPD is higher among Arab women than Jewish women who all are citizens of the state (Shwartz et al., 2019). Previous studies in Israel conducted with Arab women (including Bedouin) suggested that PPD prevalence in this population was as high as 43% (Alfayumi-Zeadna et al., 2015; Eilat-Tsanani et al., 2006; Glasser et al., 2011, 2012). Although PPD prevalence was only about half as high for Jewish women in Israel (Eilat-Tsanani et al., 2006; Glasser, 2012; Glasser et al., 1998), prevalence for both groups was greater compared with women in other countries (e.g., Brazil, Guyana, Costa Rica, Italy, Chile, South Africa, Taiwan, and Korea; Halbreich & Karkun, 2006).

Aside from hormonal factors, many other PPD risk factors have been identified. These include low education, unmarried status, unemployment, low social support (Surkan et al., 2006), high stress (Beydoun et al., 2010; Fiala et al., 2017; Gauthreaux et al., 2017; Liu et al., 2016; Mitra et al., 2015), not breastfeeding, young age (below 18 years; Robertson et al., 2004), having a female infant (Deng et al., 2014; Liu & Tronick, 2013), economic difficulties, having other children below age 5, being an immigrant (Daoud et al., 2019; Robertson et al., 2004), and experiencing discrimination (Surkan et al., 2006). Pregnancy and childbirth-related factors that also increase PPD risk include unwanted pregnancy (Gauthreaux et al., 2017; Hayes et al., 2010; Liu & Tronick, 2013; Urquia et al., 2017) and preterm birth (Gauthreaux et al., 2017). Previous research suggests that intimate partner violence (IPV) is also a predictor for PPD (Faisal-Cury et al., 2013; Ludermir et al., 2010; Tiwari et al., 2008; Valentine et al., 2011; Velonis et al., 2017).

IPV defines a broad range of violent or aggressive actions, including physical and sexual violence, stalking, and psychological aggression (including coercive tactics), perpetrated by a person’s current or former intimate partner (spouse, boyfriend/girlfriend, dating partner, or ongoing sexual partner) (Breiding et al., 2015). IPV is a significant public health problem. For example, a global World Health Organization (WHO) study found that lifetime prevalence of IPV among women was 13% to 61% (WHO, 2013) and the 2011 National Intimate Partner and Sexual Violence Survey (NISVS) in the United States found that more than one in five women (22.3%) and nearly one in seven men (14.0%) experienced severe physical violence by an intimate partner at some point in their adult lives. NISVS data also showed that 9.2% of women and 2.5% of men had been stalked by an intimate partner (Breiding et al., 2015). According to European Union (EU) statistics, 22% of women 15+ years of age and more experienced IPV. Comparing rates across countries, IPV rates range from 13% in Austria, Croatia, Poland, Slovenia, and Spain to 30% to 32% in Finland, Denmark, and Latvia (European Union Agency for Fundamental Rights, 2014).

Men and women of all social strata experience IPV. However, incidence is higher for women and for certain social and ethnic groups. IPV risk factors include a history of violence (Coker et al., 2000; Finnbogadottir et al., 2014; Guo et al., 2004), younger age (Taillieu & Brownridge, 2010), being unmarried (Charles & Perreira, 2007; Saltzman et al., 2003), belonging to an ethnic minority group (Charles & Perreira, 2007; Janssen et al., 2003; Saltzman et al., 2003), unemployment (Bohn et al., 2004; Ludermir et al., 2010), low income (Taillieu & Brownridge, 2010), economic dependence (Fanslow et al., 2008; Perales et al., 2009), low education (Ludermir et al., 2010; Saltzman et al., 2003; Taillieu & Brownridge, 2010), spousal unemployment (Golchin et al., 2014), and stress (Capaldi et al., 2012).

Additional risk factors include unplanned or unwanted pregnancy (Tiwari et al., 2008; Urquia et al., 2017) and prior miscarriage or abortion (Finnbogadottir et al., 2014; Taillieu & Brownridge, 2010). Pregnant women are at heightened risk for IPV due to their increased physical, social, emotional, and economic vulnerability (Taillieu & Brownridge, 2010). IPV during pregnancy is particularly problematic due to adverse effects on maternal and fetal health (Janssen et al., 2003; Perales et al., 2009; Taillieu & Brownridge, 2010). For women who have experienced IPV with their current partner, violence can continue during pregnancy, can desist (Guo et al., 2004; Janssen et al., 2003; Saltzman et al., 2003), or can increase (Saltzman et al., 2003). Women who experience violence during pregnancy are at increased risk for violence during future pregnancies (Fanslow et al., 2008; Guo et al., 2004).

Low social support is a further risk factor for IPV (Ludermir et al., 2010). Violent spouses may attempt to isolate their partners and sever supportive relationships with family and friends (Bacchus et al., 2006). Indeed, social support is one protective factor against PPD as demonstrated in several studies (Coker et al., 2002; Faisal-Cury et al., 2013; Robertson et al., 2004). Across populations, lack of social support increases the likelihood of depression (Beydoun et al., 2010; Kendall-Tackett, 2007), and depression appears to impair the willingness and ability to make and maintain social connections (Kendall-Tackett, 2007). A U.S. study conducted in the first year after childbirth found an association between greater social support and reduced depression (Surkan et al., 2006). Women who reported having two or more supportive relationships (e.g., family, friends) reported fewer depressive symptoms than those who reported less social support (Surkan et al., 2006). Similarly, researchers in Brazil, who found a significant association between psychological, physical, and sexual IPV and PPD, also reported that social support was associated with reduced PPD risk for all types of IPV (Faisal-Cury et al., 2013). Furthermore, a U.S. study found that women who reported IPV but had high social support reported significantly less physical and psychological distress, anxiety, and depressive symptoms compared with women who reported low social support (Coker et al., 2002).

The association between IPV and PPD has been demonstrated in ample research (Bacchus et al., 2018; Beydoun et al., 2010, 2012; Faisal-Cury et al., 2013; Howard et al., 2013; Miura & Fujiwara, 2017). Between 9% and 28% of women with PPD report experiencing IPV at some point in their lives (Beydoun et al., 2012). Psychiatric symptoms around the time of birth are associated with IPV prior to and during pregnancy (Howard et al., 2013). Moreover, the risk of mental illness for pregnant women increases 1.5 times for each case of emotional abuse and two times for each instance of sexual abuse (Golchin et al., 2014). The higher the frequency and severity of IPV, the greater the risk for depression (Beydoun et al., 2012).

A recent meta-analysis by Bacchus and colleagues (2018) reported a significant association between IPV and PPD, with a pooled odds ratio (OR) of 2.19. In addition, depression symptom severity was associated with greater IPV risk. These findings suggest that the relationship between IPV and PPD might be bidirectional (Tsai et al., 2016). In an earlier meta-analysis, Wu and colleagues (2012) also found a positive association between IPV and PPD, with IPV emerging as a primary predictor of PPD (OR = 3.47) compared with women who did not experience IPV. In their meta-analysis, Beydoun and colleagues (2012) reported a moderate to strong positive association between IPV and PPD and a 1.5 to 2.0 greater risk of PPD for women who experienced IPV. A cohort study of U.S. women found that IPV in the 12 months before or during pregnancy led to a fivefold increase in PPD risk in the first year after childbirth (Valentine et al., 2011). Similarly, a more recent study of pregnant women in Japan found a positive dose–response effect of both physical and verbal IPV on PPD (Miura & Fujiwara, 2017).

To date, however, research examining the pathways between IPV and PPD is sparse. There are also comparatively few multivariate studies that include predictors, such as social support, or other sociodemographic, socioeconomic, stress-related, or obstetric health factors. As well, few studies have considered associations between IPV and PPD across ethnic groups. Our previous research showed higher PPD (Shwartz et al., 2019) and IPV (Daoud et al., 2017) among Arab women compared with Jewish women. For this study, we set out to identify both direct and indirect predictors of PPD, including IPV, in a national sample of Jewish and Arab women of childbearing age in Israel.

Method

Data for this study are part of a larger cohort study of “Family relations, violence and health” (Daoud et al., 2017), which was approved by the Review Ethics Board at Ben-Gurion University of the Negev (BGU) and the Public Health Division of the Israeli Ministry of Health (MCH). Data were collected between October 2014 and October 2015.

Study Design and Sampling

This study sample comprised 1,055 women including 807 Jewish and 248 Arab who were 6 weeks to 6 months postpartum and participated in the study on “Family Relations, Violence and Health” conducted in 2014–2015 (Daoud et al., 2017). The original study included 1,401 women citizens in Israel who were pregnant or 6 weeks to 6 months postpartum, of whom two thirds were Jewish women and one third were Arab women commensurate with the ethnic composition of women at the reproductive age. We included only postpartum women in this study, and the ratio of Jewish to Arab women was changed.

Trained female interviewers conducted interviews with participants (aged 18–48 years at that time) in private rooms, using equivalent Hebrew and Arabic structured questionnaires, when attending one of 63 maternal and child health (MCH) clinics across five health districts in Israel: Central, South, Ashkelon, North, and Haifa. Each district includes MCH clinics across diverse communities, including economically poor, average, and more affluent communities, to enable data collection across socioeconomic and ethnic groups. The study team first met with regional physicians and nurses to describe and explain the study to foster data collection. Each district consists of a number of subdistricts, and we made contact with subdistrict nurses to assist in locating clinics in which to collect data. The response rate from the larger cohort study was 73% among Jewish women and 76% among Arab women (authors’ names withheld).

Study Measures

PPD

PPD was measured using the Edinburgh Postnatal Depression Scale (EPDS). The EPDS consists of 10 questions about a woman’s feelings and mood in the last 2 weeks, such as “I have felt worried and anxious for no good reason” and “In the past week I felt sad or unhappy (moody),” with four response alternatives (0–3), and a final question pertaining to thoughts of self-harm (i.e., suicide; Cox et al., 1987). All women who endorsed this question in the positive were referred for clinical assessment.

All women were screened for PPD by the MCH clinic nurses before our interview. Internal consistency of responses to the English version of the EPDS is high (α = .87; Cox et al., 1987). Responses greater than 10 are suggestive of clinically significant PPD (Cox et al., 1987; Murray & Carothers, 1990). At this level, the Hebrew version of the EPDS was found to have 92.3% sensitivity and 72.5% specificity (Kandel-Katznelson et al., 2000). Responses to the Arabic version of the EPDS have similar sensitivity, at 91% and 84% specificity for PPD detection (Ghubash et al., 1997). For this study sample, internal consistency was again high (α = .82). In our analysis, PPD was examined as a continuous variable.

IPV recurrence

IPV was measured by 10 self-report questions used in a previous research in Israel (Daoud et al., 2017). The questionnaire measures the frequency of physical, sexual, emotional, social, and economic IPV (e.g., “Are you afraid of extreme changes in your partner’s mood?”; “Is your partner trying to isolate you from family and friends?”; “Did your partner hit you, kick you, push you, or throw objects at you?”), with four response categories ranging from never (0) to always (3).

Social support

Social support was measured by six questions regarding emotional, physical, or material support, as well as information about support resources (i.e., “Is there someone who will give you information, guidance or advice during a crisis?”; “Is there someone you can trust, talk to about yourself or your problems?”; “Is there someone who supports you emotionally during a crisis?”; “Is there someone who will give you material help, for example, lend you money if you need it?”). Responses ranged from never (0) to always (4) (Daoud et al., 2009a, 2009b). In previous research with Arabs in Israel, internal consistency of responses was high (α = .87; Daoud et al., 2009b). In the current sample, α = .85 for Jewish women and α = .88 for Arab women.

Chronic stress

Chronic stress was measured by 10 yes/no questions about financial, social, family, and work problems (e.g., death in the family in the last year). This variable was used in previous research with Jewish and Arab women in Israel (Daoud et al., 2009b, 2018).

Sociodemographics

The sociodemographic questions included ethnicity (Jewish or Arab), educational attainment (high school or below, above high school, or university/college degree), and current employment status (full time, part time, or unemployed). Finally, we asked if the participant’s most recent pregnancy was planned or unplanned.

Statistical Analysis

Path analysis is an extension of linear regression with significant advantages (O’Rourke & Hatcher, 2013); for instance, path analysis allows us to simultaneously predict one or more dependent variables (touched by an arrowhead in path models). Arrows pointing from independent to dependent variables represent significant prediction (i.e., critical ratio values > |1.96|). Path analysis is a multivariate statistical procedure, meaning that all significant paths emerged concurrently (i.e., over and above other statistically significant results). Preliminary analyses indicated that chronic stress and ethnicity moderate or mediate the association between IPV and PPD. This result informed the baseline path analytic model we computed next.

Path analysis allows us to identify both direct and indirect predictors of PPD. Indirect prediction occurs via other variables (i.e., two or more pathways between variables). In complex or more nuanced path models, variables can have direct and indirect effects on dependent variables, and indirect effects can be of equal or greater magnitude than direct effects (total effects = direct effects + indirect effects).

Computing path analyses with the SEM software allowed us to obtain overall goodness-of-fit information, and good model fit is required to interpret individual results. Consistent with the SEM best practices, we report three goodness-of-fit indices: an incremental (e.g., CFI [confirmatory fit index]), an absolute (e.g., SRMR [standardized root mean square residual]), and a parsimonious (e.g., RMSEA [root mean square error of approximation]) fit index.

For this study, we performed path analyses in a three-step process, as described by O’Rourke and Hatcher (2013). We first computed a baseline model in which all independent variables were assumed to directly predict PPD. We assumed that stress and other sociodemographic variables predicted both IPV recurrence and PPD (model not shown here) among Arab and Jewish women. Next, we deleted from the model all hypothesized nonsignificant paths from the baseline model. Finally, we added unhypothesized but statistically significant paths where justified based on theory or prior research. With eight independent variables, our combined sample of 275 Arab and 853 Jewish women was sufficient to detect medium to small effect sizes (α = .01). We computed path models using AMOS version 24.

Results

A total of 1,055 women were included in our analysis, of whom 76.5% were Jewish and 23.5% were Arab. Almost half the women had an academic degree, and two thirds worked full/part time. About 10% reported that their pregnancy was not planned. Table 1 reports descriptive statistics.

Table 1.

Descriptive Statistics for the Study Variables.

Variables	Total Sample (N = 1,055)
Ethnicity, n (%)
Jewish	807 (76.5)
Arab	248 (23.5)
Women’s education, n (%)
High school and below	346 (32.8)
College above high school	190 (18.0)
Academic degree	519 (49.2)
Current employment, n (%)
Full or part time	655 (62.1)
Not working	400 (37.9)
Planned pregnancy, n (%)
No	108 (10.2)
Yes	947 (89.8)
EPDS (PPD), M (SD) range	3.65 (4.34) 0–23
Intimate partner violence, M (SD) range	0.67 (1.18) 0–7
Social support, M (SD) range	4.53 (0.77) 1–5
Chronic stress, M (SD) range	1.92 (1.81) 0–9

Note. EPDS = Edinburgh Postnatal Depression Scale; PPD = postpartum depression.

About 10% of women reported responses suggestive of PPD. However, the percentage was three times greater for Arab (20.7%) than Jewish (7%) women (shown elsewhere).

Table 2 shows the correlations between the study variables. Statistically significant correlations were found between EPDS and all study variables (p < .01). Positive correlation coefficients (r) were found between EPDS and IPV recurrence (r = .283), chronic stress (r = .413), unplanned pregnancy (r = .093), and ethnicity (r = .125). Negative correlations were found with women’s education (r = −.217) and current employment (r = −.111).

Table 2.

Correlations Between the Study Variables.

Variables	1	2	3	4	5	6	7	8
1. EPDS (PPD)	1.000	.283**	.413**	−.217**	0.059	−.111**	.093**	.125**
2. Intimate partner violence		1.000	.314**	−.313**	.175**	−.185**	.180**	.320**
3. Chronic stress			1.000	−.169**	−.020	.015	.090**	.049
4. Social support				1.000	−.186**	.227**	−.121**	−.303**
5. Women’s education					1.000	−.317**	.115**	.277**
6. Current employment						1.000	−.078*	−.378**
7. Unplanned pregnancy							1.000	.167**
8. Ethnicity								1.000

Note. N = 1,055 for all correlations. EPDS = Edinburgh Postnatal Depression Scale; PPD = postpartum depression.

Correlation is significant at the .05 level (two tailed). **Correlation is significant at the .01 level (two tailed).

We next performed path analyses to identify both direct and indirect predictors of PPD, including IPV. This allowed us to examine the role of ethnicity in relation to both IPV recurrence and PPD. With 1,055 participants and nine degrees of freedom, statistical power for this model was greater than 0.99 (MacCallum et al., 1996) (Figure 1).

Figure 1.

Path model predicting postpartum depression (PPD) among Jewish and Arab women in Israel (N = 1,055).

Goodness of Fit

The resulting model showed multiple direct and indirect predictors of PPD in this study sample, χ²(df = 9) = 21.05, p = .01. Goodness of fit was in ideal parameters for the CFI (CFI > .94; CFI = .99), the SRMR (SRMR > .055; SRMR = .020), and the RMSEA (RMSEA > .055; RMSEA = .036). The full 90% confidence interval for the RMSEA statistic was in the acceptable parameters (.016 < RMSEA CL₉₀ < .056).

Predictors of PPD

Five variables emerged as direct predictors of PPD (R² = .22, p < .01). Of these, the direct effect of chronic stress on PPD (β = .34) was greater than that on IPV (β = .14), planned versus unplanned pregnancy (β = .06), and social support (β = −.10) (Figure 1). Note also that the effect of IPV was both direct and indirect (i.e., via chronic stress and social support). As reported in Table 3, the indirect effect of IPV on PPD was equivalent to the direct effect. The direct and indirect effects of ethnicity on PPD (via IPV and social support) were also equivalent.

Table 3.

Direct and Indirect Predictors of Postpartum Depression for Arab and Jewish Women in Israel (Standardized Coefficients).

Variables	Intimate partner violence recurrence	Chronic stress	Social support	Postpartum depression
Ethnicity
d:	.24	.00	−.22	.08
i:	.00	.08	−.05	.09
t:	.24	.08	−.27	.17
Unplanned pregnancy
d:	.16	.00	.00	.06
i:	.00	.05	−.03	.04
t:	.16	.05	−.03	.10
Current employment
d:			.09	.00
i:			.00	−.01
t:			.09	−.01
Women’s education
d:	−.13	.00	.11	.00
i:	.00	−.04	.03	−.05
t:	−.13	−.04	.14	−.05
Intimate partner violence recurrence
d:		.32	−.20	.14
i:		.00	.00	.13
t:		.32	−.20	.27
Chronic stress
d:				.34
i:				.00
t:				.34
Social support
d:				−.10
i:				.00
t:				−.10

Note. d = direct effects; i = indirect effects; t = total effects.

Predictors of IPV

In this model, IPV was predicted by Arab ethnicity (β = .24), unplanned pregnancy (β = .16), and lower education (β = −.13). In turn, IPV predicted lower social support (β = −.20), greater chronic stress (β = .32), and PPD (β = .14) (Figure 1).

The model that emerged from these analyses indicated that many factors predict PPD. Both ethnicity and IPV emerged as direct and indirect risk factors of PPD in this sample. Of note, the buffering effect of social support was minimal relative to other variables, and the effects of social support were lower still for Arab women who reported greater IPV and PPD compared with Jewish women (Table 4).

Table 4.

Covariance Estimates Between Independent Variables Predicting Postpartum Depression Among Jewish and Arab Women in Israel.

Covariance estimates	Coefficient	Critical Ratio
Women’s education–current employment	.32	9.88**
Current employment–ethnicity	−.38	−11.47**
Unplanned pregnancy–ethnicity	.17	5.34**
Education–unplanned pregnancy	−.12	−3.82**
Women’s education–ethnicity	−.28	−8.76**
Current employment–unplanned pregnancy	−.08	−2.51*

p < 0.05. **p < .01.

Discussion

Our study results suggest that the direct and indirect effects of IPV on PPD were similar in our study sample. In other words, IPV is a significant direct predictor of PPD, with equivalent indirect effects via chronic stress and social support. We also found that ethnicity and unplanned pregnancy affect both IPV and PPD. This aligns with findings by Romito and colleagues (2009), who showed that women who experienced IPV reported more depressive symptoms after birth. They also found an indirect link between economic problems and a history of violence with psychological distress that was mediated by IPV after birth (Romito et al., 2009). Likewise, our multivariate analyses enabled us to identify several concomitant pathways by which IPV frequency increased the likelihood of PPD.

In our study, social support appeared to be related to both IPV and PPD. Women who reported higher IPV also reported less social support. Women who reported higher social support reported less PPD. Previous research examining the association between IPV and PPD has found that social support was a protective factor against PPD, reducing the risk of PPD, including among women who reported IPV (Beydoun et al., 2010; Coker et al., 2002; Faisal-Cury et al., 2013). In the United States, Coker and colleagues (2002) examined IPV and the frequency of various mental health outcomes and found that women who experienced IPV and high social support reported significantly fewer mental and physical health problems compared with women who experienced IPV and had low social support. In addition, these authors reported that women who received consistent social support were significantly less likely to experience anxiety and depression (Coker et al., 2002). Social support is a vital variable among women who are victims of IPV in the postpartum period.

We also found that chronic stress was associated with both IPV and PPD. This is consistent with previous research results, including Coker and colleagues (2002), who found that IPV indirectly affected women’s mental health by increasing psychological stress. These authors suggest that raising IPV awareness is not enough; instead, victims of IPV need emotional support to sustain their mental health (Coker et al., 2002). Fostering social support (e.g., reestablishing lost contacts) is integral to reducing the negative effects of IPV.

In their study examining the role of stress during pregnancy, LaCoursière and colleagues (2012) found that women who reported IPV also reported high frequency of stressful life events. They also found that the frequency of carcinogenic events increased PPD. In other words, the more stress women experienced, the more likely they were to report PPD (LaCoursière et al., 2012). Findings from this and previous studies (Beydoun et al., 2010; Velonis et al., 2017) indicated that women reporting IPV experienced higher levels of chronic stress, and that chronic stress was associated with increased likelihood of PPD.

We found that ethnicity both directly and indirectly predicted PPD (via IPV and social support). Similar findings were reported in a Hawaiian study including Pacific Island and Asian women (N = 7,154), which found that ethnic minority women were twice as likely to report PPD as Caucasian women (Hayes et al., 2010). Those same researchers found that ethnic minority women were more likely to experience IPV, smoke, use illegal drugs during pregnancy, and receive nutritional support as part of their prenatal care (Hayes et al., 2010). Elsewhere, compared with nonimmigrant women, Indigenous and immigrant women had higher PPD (Daoud et al., 2019) and higher IPV, with lower social support (Daoud et al., 2013).

Arab women recruited for this study reported high frequency of IPV and high PPD independent of other sociodemographic factors (e.g., education, employment). This finding underscores the need for culturally appropriate interventions developed for and with Arab communities to help identify and reduce IPV (Daoud et al., 2017). Our findings suggest that this may also reduce PPD for this population.

We also found that women who did not plan their pregnancy and reported higher IPV were at greater PPD risk. This finding is also consistent with existing research (Gauthreaux et al., 2017; Urquia et al., 2017). One U.S. study found that, compared with women who did plan their pregnancy, women who did not plan their pregnancy were more likely to experience symptoms of PPD. These women were more likely to be unmarried and to have low income, limited education, a history of depression, a history of abuse, and other life stressors (Gauthreaux et al., 2017). Meanwhile, a study by Urquia and colleagues (2017) found that women whose partners did not want the pregnancy were at higher risk of IPV and PPD. It is possible that women whose pregnancy was not planned are at higher PPD risk, in part because they receive less support from disgruntled partners, making it much harder to cope with the pregnancy and the baby (Urquia et al., 2017).

Limitations and Strengths of the Study

This was a cross-sectional study. It is not possible to conclude that significant associations between IPV, PPD, and other variables are causal. Path analysis allowed us to determine the likelihood of the direction of associations between variables (i.e., prediction), but these findings need to be replicated in longitudinal research and with other samples. The second limitation was that the questionnaire took about 20 to 30 min to complete, which might have deterred some women attending the MCH clinics from participating due to lack of time or a crying baby, post-vaccination. This could have caused some selection bias, as women who experience IPV or PPD might have more often refused participation than women with no experiences of IPV or PPD as they might fear the consequences of disclosing IPV to the study team, if the abusive partner knows about her participation in the study. Still, about 40% of women in the study disclosed IPV, and PPD was close to 10%. We made an effort to reduce bias by instructing our interviewers to ask women who declined being interviewed during the MCH visit if they would consider being interviewed on another visit.

The main strength of our study was the large sample size. The study was conducted with Jewish and Arab women, postpartum, in all health districts in Israel in MCH clinics serving high-, medium-, and low-income communities. Pre- and postnatal care is provided under Israel’s National Health Insurance Law, meaning that the findings of the study can likely be generalized to the population with some confidence. Also, this is the first study conducted in Israel to examine the relationship between IPV and PPD among women of childbearing age, and among the few to have conducted path analysis to explore direct and indirect pathways linking IPV to PPD.

Conclusion

Our findings indicate that IPV frequency directly and indirectly predicts PPD, mediated by ethnicity, chronic stress, and social support. When comparing subgroups in our Israeli sample, Arab women were at increased risk for IPV, lower social support, and higher PPD than Jewish women. These results underscore the need for culturally sensitive programs to identify and reduce IPV and foster social support. Such programs may reduce the feelings of chronic stress reported by women and indirectly help reduce PPD. Lowering rates of violence will also help reduce the disparity in PPD between Jewish and Arab women in Israel.

Footnotes

Acknowledgments

The authors thank the women who agreed to participate in the study.

Author Contributions

N.D. conceived of the study, was the primary investigator, and supervised the data collection, data cleaning, and analysis. N.S., N.O., and N.D. conducted the data analyses, wrote the manuscript, and approved the final version before submission. This paper has not been previously published. All authors approved its publication.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by the Israel Science Foundation (ISF; grant no. 881/13). The funding agency had no role in the study. The study reflects the work of the coauthors.

Ethical approval

This study was approved by the Ethics Review Board of Ben-Gurion University of the Negev (BGU) and the Public Health Division at the Israeli Ministry of Health (MCH).

ORCID iD

Nihaya Daoud

Author Biographies

Nitza Shwartz, RN, MPH, is a graduate Master degree at Ben-Gurion University of the Negev. Her thesis research focuses on investigating unique characteristics in Arab and Jewish women’s populations in Israel, with a special interest in intimate partner violence and postpartum depression.

Norm O’Rourke, PhD, is a professor and clinical psychologist at Ben-Gurion University of the Negev. He is the founding director of the Foundation for Wellness with Bipolar Disorder, was elected to the executive board of APA’s Society for Clinical Geropsychology, and sat on the Mental Health Commission of Canada. He is currently an associate editor of Aging and Mental Health.

Nihaya Daoud, PhD, is a social epidemiologist and health promotion specialist at the School of Public Health in Ben-Gurion University of the Negev, Israel. She is the founder and director of the Center for Health Equity and Health promotion at the Faculty of Health Sciences. Her research focuses on health inequities and intersectionality of gender, class, and ethnicity among minorities and marginalized populations. Her recent research explores health and health care services for women experiencing IPV, and how masculinity interacts with violence against women in Israel and other EU countries.

References

Alfayumi-Zeadna

Kaufman-Shriqui

Zeadna

Lauden

Shoham-Vardi

(2015). The association between sociodemographic characteristics postpartum depression symptoms among Arab-Bedouin women in southern Israel. Depression and Anxiety, 32(2), 120–128.

Bacchus

L. J.

Mezey

Bewley

(2006). A qualitative exploration of the nature of domestic violence in pregnancy. Violence Against Women, 12(6), 588–604. https://doi.org/10.1177/1077801206289131

Bacchus

L. J.

Ranganathan

Watts

Devries

(2018). Recent intimate partner violence against women and health: A systematic review and meta-analysis of cohort studies. BMJ Open, 8(7), Article e019995.

Beydoun

H. A.

Al-Sahab

Beydoun

M. A.

Tamim

(2010). Intimate partner violence as a risk factor for postpartum depression among Canadian women in the Maternity Experience Survey. Annals of Epidemiology, 20(8), 575–583.

Beydoun

H. A.

Beydoun

M. A.

Kaufman

J. S.

Zonderman

A. B.

(2012). Intimate partner violence against adult women and its association with major depressive disorder, depressive symptoms and postpartum depression: A systematic review and meta-analysis. Social Science & Medicine, 75(6), 959–975.

Bohn

D. K.

Tebben

J. G.

Campbell

J. C.

(2004). Influences of income, education, age, and ethnicity on physical abuse before and during pregnancy. Journal of Obstetric, Gynecologic, & Neonatal Nursing, 33(5), 561–571.

Breiding

Basile

K. C.

Smith

S. G.

Black

M. C.

Mahendra

R. R.

(2015). Intimate partner violence surveillance: Uniform definitions and recommended data elements (Version 2.0). National Center for Injury Prevention and Control, Centers for Disease Control and Prevention.

Capaldi

D. M.

Knoble

N. B.

Shortt

J. W.

Kim

H. K.

(2012). A systematic review of risk factors for intimate partner violence. Partner Abuse, 3(2), 231–280.

Ceballos

Wallace

Goodwin

(2016). Postpartum depression among African-American and Latina mothers living in small cities, towns, and rural communities. Journal of Racial and Ethnic Health Disparities, 4(5), 916–927.

10.

Charles

Perreira

K. M.

(2007). Intimate partner violence during pregnancy and 1-year post-partum. Journal of Family Violence, 22(7), 609–619.

11.

Coker

A. L.

Smith

P. H.

McKeown

R. E.

King

M. J.

(2000). Frequency and correlates of intimate partner violence by type: Physical, sexual, and psychological battering. American Journal of Public Health, 90(4), 553–559.

12.

Coker

A. L.

Smith

P. H.

Thompson

M. P.

McKeown

R. E.

Bethea

Davis

K. E.

(2002). Social support protects against the negative effects of partner violence on mental health. Journal of Women’s Health & Gender-Based Medicine, 11(5), 465–476.

13.

Cox

J. L.

Holden

J. M.

Sagovsky

(1987). Detection of postnatal depression: Development of the 10-item Edinburgh Postnatal Depression Scale. The British Journal of Psychiatry, 150(6), 782–786.

14.

Daoud

O’Brien

O’Campo

Harney

Bebee

… Smylie

(2019). Postpartum depression prevalence and risk factors among Indigenous, non-Indigenous and immigrant women in Canada. Canadian Journal of Public Health, 110, 440–452.

15.

Daoud

Sergienko

Shoham-Vardi

(2017). Intimate partner violence prevalence, recurrence, types, and risk factors among Arab, and Jewish immigrant and nonimmigrant women of childbearing age in Israel. Journal of Interpersonal Violence. Advance online publication. https://doi.org/10.1177/0886260517705665

16.

Daoud

Smylie

Urquia

Allan

O’Campo

(2013). The contribution of socio-economic position to the excesses of violence and intimate partner violence among Aboriginal versus non-Aboriginal women in Canada. Canadian Journal of Public Health, 104(4), e278–e283.

17.

Daoud

Soskolne

Manor

(2009a). Educational inequalities in self-rated health within the Arab minority in Israel: Explanatory factors. The European Journal of Public Health, 19(5), 477–483.

18.

Daoud

Soskolne

Manor

(2009b). Examining cultural, psychosocial, community and behavioural factors in relationship to socioeconomic inequalities in limiting longstanding illness among the Arab minority in Israel. Journal of Epidemiology and Community Health, 63(5), 351–358.

19.

Daoud

Soskolne

Mindell

J. S.

Roth

M. A.

Manor

(2018). Ethnic inequalities in health between Arabs and Jews in Israel: The relative contribution of individual-level factors and the living environment. International Journal of Public Health, 63(3), 313–323.

20.

Deng

A.-W.

Xiong

R.-B.

Jiang

T.-T.

Luo

Y.-P.

Chen

W.-Z.

(2014). Prevalence and risk factors of postpartum depression in a population-based sample of women in Tangxia Community, Guangzhou. Asian Pacific Journal of Tropical Medicine, 7(3), 244–249.

21.

Eilat-Tsanani

Merom

Romano

Reshef

Lavi

Tabenkin

(2006). The effect of postpartum depression on women’s consultations with physicians. The Israel Medical Association Journal, 8(6), 406–410.

22.

European Union Agency for Fundamental Rights. (2014). Violence against women: An EU-wide survey 2014. Publications Office of the European Union.

23.

Faisal-Cury

Menezes

P. R.

d’Oliveira

A. F. P. L.

Schraiber

L. B.

Lopes

C. S.

(2013). Temporal relationship between intimate partner violence and postpartum depression in a sample of low income women. Maternal and Child Health Journal, 17(7), 1297–1303.

24.

Fanslow

Silva

Whitehead

Robinson

(2008). Pregnancy outcomes and intimate partner violence in New Zealand. Australian and New Zealand Journal of Obstetrics and Gynaecology, 48(4), 391–397.

25.

Fiala

Švancara

Klánová

Kašpárek

(2017). Sociodemographic and delivery risk factors for developing postpartum depression in a sample of 3233 mothers from the Czech ELSPAC study. BMC Psychiatry, 17(1), Article 104.

26.

Finnbogadottir

Dykes

A. K.

Wann-Hansson

(2014). Prevalence of domestic violence during pregnancy and related risk factors: A cross-sectional study in southern Sweden. BMC Women’s Health, 14, Article63. https://doi.org/10.1186/1472-6874-14-63

27.

Gauthreaux

Negron

Castellanos

Ward-Peterson

Castro

de la Vega

P. R.

Acuña

J. M.

(2017). The association between pregnancy intendedness and experiencing symptoms of postpartum depression among new mothers in the United States, 2009 to 2011: A secondary analysis of PRAMS data. Medicine, 96(6), Article e5851.

28.

Ghubash

Abou-Saleh

Daradkeh

(1997). The validity of the Arabic Edinburgh Postnatal Depression Scale. Social Psychiatry and Psychiatric Epidemiology, 32(8), 474–476.

29.

Glasser

(2012). Postpartum depression: Early detection of primary health services in Israel. http://www.hebpsy.net/articles.asp?id=2772

30.

Glasser

Barell

Shoham

Ziv

Boyko

Lusky

Hart

(1998). Prospective study of postpartum depression in an Israeli cohort: Prevalence, incidence and demographic risk factors. Journal of Psychosomatic Obstetrics & Gynecology, 19(3), 155–164.

31.

Glasser

Stoski

Kneler

Magnezi

(2011). Postpartum depression among Israeli Bedouin women. Archives of Women’s Mental Health, 14(3), 203–208.

32.

Glasser

Tanous

Shihab

Goldman

Ziv

Kaplan

(2012). Perinatal depressive symptoms among Arab women in northern Israel. Maternal and Child Health Journal, 16(6), 1197–1205.

33.

Golchin

N. A. H.

Hamzehgardeshi

Ahoodashti

M. S.

(2014). Sociodemographic characteristics of pregnant women exposed to domestic violence during pregnancy in an Iranian setting. Iranian Red Crescent Medical Journal, 16(4), Article e11989. https://doi.org/ARTNe1198910.5812/ircmj.11989

34.

Guo

Yan

(2004). Physical and sexual abuse of women before, during, and after pregnancy. International Journal of Gynecology & Obstetrics, 84(3), 281–286.

35.

Halbreich

Karkun

(2006). Cross-cultural and social diversity of prevalence of postpartum depression and depressive symptoms. Journal of Affective Disorders, 91(2), 97–111.

36.

Haque

Namavar

Breene

K.-A.

(2015). Prevalence and risk factors of postpartum depression in Middle Eastern/Arab women. Journal of Muslim Mental Health, 9(1), 65–84.

37.

Hayes

D. K.

V. M.

Hurwitz

E. L.

Mitchell-Box

K. M.

Fuddy

L. J.

(2010). Disparities in self-reported postpartum depression among Asian, Hawaiian, and Pacific Islander Women in Hawaii: Pregnancy Risk Assessment Monitoring System (PRAMS), 2004–2007. Maternal and Child Health Journal, 14(5), 765–773.

38.

Howard

L. M.

Oram

Galley

Trevillion

Feder

(2013). Domestic violence and perinatal mental disorders: A systematic review and meta-analysis. PLOS Medicine, 10(5), Article e1001452.

39.

Janssen

P. A.

Holt

V. L.

Sugg

N. K.

Emanuel

Critchlow

C. M.

Henderson

A. D.

(2003). Intimate partner violence and adverse pregnancy outcomes: A population-based study. American Journal of Obstetrics and Gynecology, 188(5), 1341–1347.

40.

Kandel-Katznelson

Maisel

Zilber

Lerner

(2000). Validation of the Hebrew translation of the Edinburgh Postnatal Depression Scale: Background, methods and findings [Conference session]. Tenth Annual Conference of the Israeli Psychiatric Association, Jerusalem, Hebrew.

41.

Kendall-Tackett

K. A.

(2007). Violence against women and the perinatal period the impact of lifetime violence and abuse on pregnancy, postpartum, and breastfeeding. Trauma, Violence, & Abuse, 8(3), 344–353.

42.

LaCoursière

D. Y.

Hirst

K. P.

Barrett-Connor

(2012). Depression and pregnancy stressors affect the association between abuse and postpartum depression. Maternal and Child Health Journal, 16(4), 929–935.

43.

Liu

C. H.

Giallo

Doan

S. N.

Seidman

L. J.

Tronick

(2016). Racial and ethnic differences in prenatal life stress and postpartum depression symptoms. Archives of Psychiatric Nursing, 30(1), 7–12.

44.

Liu

C. H.

Tronick

(2013). Rates and predictors of postpartum depression by race and ethnicity: Results from the 2004 to 2007 New York City PRAMS survey (Pregnancy Risk Assessment Monitoring System). Maternal and Child Health Journal, 17(9), 1599–1610.

45.

Ludermir

A. B.

Lewis

Valongueiro

S. A.

de Araújo

T. V. B.

Araya

(2010). Violence against women by their intimate partner during pregnancy and postnatal depression: A prospective cohort study. The Lancet, 376(9744), 903–910.

46.

MacCallum

R. C.

Browne

M. W.

Sugawara

H. M.

(1996). Power analysis and determination of sample size for covariance structure modeling. Psychological Methods, 1(2), 130–149.

47.

Mitra

Iezzoni

L. I.

Zhang

Long-Bellil

L. M.

Smeltzer

S. C.

Barton

B. A.

(2015). Prevalence and risk factors for postpartum depression symptoms among women with disabilities. Maternal and Child Health Journal, 19(2), 362–372.

48.

Miura

Fujiwara

(2017). Intimate partner violence during pregnancy and postpartum depression in Japan: A cross-sectional study. Frontiers in Public Health, 5, Article 81. https://doi.org/10.3389/fpubh.2017.00081

49.

Murray

Carothers

A. D.

(1990). The validation of the Edinburgh Post-natal Depression Scale on a community sample. The British Journal of Psychiatry, 157(2), 288–290.

50.

O’Rourke

Hatcher

(2013). A step-by-step approach to using SAS for factor analysis and structural equation modeling. SAS Institute.

51.

Perales

M. T.

Cripe

S. M.

Lam

Sanchez

S. E.

Sanchez

Williams

M. A.

(2009). Prevalence, types, and pattern of intimate partner violence among pregnant women in Lima, Peru. Violence Against Women, 15(2), 224–250.

52.

Robertson

Grace

Wallington

Stewart

D. E.

(2004). Antenatal risk factors for postpartum depression: A synthesis of recent literature. General Hospital Psychiatry, 26(4), 289–295.

53.

Romito

Turan

J. M.

Neilands

Lucchetta

Pomicino

Scrimin

(2009). Violence and women’s psychological distress after birth: An exploratory study in Italy. Health Care for Women International, 30(1–2), 160–180.

54.

Saltzman

L. E.

Johnson

C. H.

Gilbert

B. C.

Goodwin

M. M.

(2003). Physical abuse around the time of pregnancy: An examination of prevalence and risk factors in 16 states. Maternal and Child Health Journal, 7(1), 31–43.

55.

Shwartz

Shoahm-Vardi

Daoud

(2019). Postpartum depression among Arab and Jewish women in Israel: Ethnic inequalities and risk factors. Midwifery, 70, 54–63.

56.

Surkan

P. J.

Peterson

K. E.

Hughes

M. D.

Gottlieb

B. R.

(2006). The role of social networks and support in postpartum women’s depression: A multiethnic urban sample. Maternal and Child Health Journal, 10(4), 375–383.

57.

Taillieu

T. L.

Brownridge

D. A.

(2010). Violence against pregnant women: Prevalence, patterns, risk factors, theories, and directions for future research. Aggression and Violent Behavior, 15(1), 14–35.

58.

Tiwari

Chan

K. L.

Fong

Leung

Brownridge

D. A.

Lam

… Chan

(2008). The impact of psychological abuse by an intimate partner on the mental health of pregnant women. BJOG: An International Journal of Obstetrics & Gynaecology, 115(3), 377–384.

59.

Tsai

A. C.

Tomlinson

Comulada

W. S.

Rotheram-Borus

M. J.

(2016). Intimate partner violence and depression symptom severity among South African women during pregnancy and postpartum: Population-based prospective cohort study. PLOS Medicine, 13(1), Article e1001943.

60.

Urquia

M. L.

Pulver

Heaman

M. I.

Ray

J. G.

Daoud

O’Campo

(2017). Partner disengagement from pregnancy and adverse maternal and infant outcomes. Journal of Women’s Health, 26(3), 234–240.

61.

Valentine

J. M.

Rodriguez

M. A.

Lapeyrouse

L. M.

Zhang

(2011). Recent intimate partner violence as a prenatal predictor of maternal depression in the first year postpartum among Latinas. Archives of Women’s Mental Health, 14(2), 135–143.

62.

Velonis

A. J.

O’Campo

Kaufman-Shriqui

Kenny

Schafer

Vance

… Chinchilli

V. M.

(2017). The impact of prenatal and postpartum partner violence on maternal mental health: Results from the Community Child Health Network Multisite Study. Journal of Women’s Health, 26, 1053–1061.

63.

World Health Organization. (2013). Responding to intimate partner violence and sexual violence against women: WHO clinical and policy guidelines.

64.

Chen

H.-L.

X.-J.

(2012). Violence as a risk factor for postpartum depression in mothers: A meta-analysis. Archives of Women’s Mental Health, 15(2), 107–114.