The contribution of intimate partner violence in exacerbating health inequalities between Palestinian and Jewish women in Israel

Abstract

Background

Across the globe, Indigenous women in subordinate social positions (minoritized, racialized, facing discrimination) experience poorer health outcomes and greater social inequalities than non-Indigenous women. Although intimate partner violence (IPV) may significantly exacerbate these disparities, the extent of IPVs contribution to the excess of health inequality has not been systematically quantified.

Methods

We estimated IPVs relative contribution to the excess of health inequalities between Indigenous Palestinian (N = 436) and Jewish women (N = 965) citizens in Israel, aged 18–50. We calculated adjusted odds ratios (AORs) and 95% confidence intervals for 10 mental and physical health conditions, considering socioeconomic and demographic factors. Mental health outcomes included: postpartum depression (PPD; EPDS ≥ 10), depressive symptoms (CES > 0.9), anxiety (STAI ≥14), self-rated health (SRH). Physical health included: abortions, miscarriages, preterm birth, unplanned pregnancy, chronic illness. Multimorbidity included: two-plus above conditions. IPVs specific contribution to health inequalities was calculated by % change (Δ) in AORs from adjusted model (socioeconomics and demographics) to a model that also considered IPV.

Results

Palestinian women had significantly poorer health than Jewish women for 7 of 10 conditions: OR (95%CI): PPD = 3.57 (2.41–5.31); depression = 3.46 (2.40–4.99); anxiety = 2.06 (1.63–2.60); unplanned pregnancy = 4.83 (3.18–7.34); miscarriages = 1.89 (1.47–2.71); preterm birth = 1.97 (1.51–2.57); multimorbidity = 1.48 (1.10–1.99). OR for chronic illness was significantly lower among Palestinian women (0.11, 0.05–0.23). Abortions and SRH were non-significant. Adjusting for IPV above socioeconomic and demographic factors, AORs for ethnonational inequalities were attenuated: physical health, 0.09% to 50.4%; mental health, 23.3% to 57.8%; multimorbidity, 29.7% (non-significant). The net contribution of IPV to the excess of ethnonational health inequalities was as follows: depression = 16.76%, anxiety = 20.39%, PPD = 20.19%, SRH = 14.65%, chronic illness = no contribution, abortions = 2.60%, miscarriages = 3.17%, preterm birth = 4.09, planned pregnancy = 24.43%, and multimorbidity = 24.33%.

Conclusions

Intimate partner violence (IPV) is a structural phenomenon shaped by intersecting social, economic, and systemic determinants that intensify health disparities between Palestinian and Jewish women during pregnancy and the postpartum period, and beyond what can be explained by socioeconomic or sociodemographic factors alone. Effectively addressing IPV and its underlying structural and social causes is essential for mitigating these persistent health inequalities.

Background

Palestinian Arab citizens of Israel (hereafter, Palestinians) constitute an Indigenous population living within a deeply contested sociopolitical and historical context shaped by ongoing conflict, displacement, and structural inequality. After the 1948 Nakba, they became a minoritized group in the Jewish-majority state [1]. Palestinian citizens make up approximately 20% of Israel’s population and reside predominantly in geographically segregated, under-resourced areas consistently ranked low for socioeconomic indicators [2]. Decades of structurally embedded discriminatory policies have led to persistent disparities in Palestinians’ access to healthcare and social services, and public resources [3]. Despite Palestinians’ historical and cultural ties to the land, international recognition of them as an Indigenous people has lagged [1], contributing to their exclusion from global Indigenous health research and a failure to acknowledge the sociopolitical context shaping their health in Israel [4]. Consequently, the distinct and disproportionate health inequalities they experience often go unaddressed [5]. Despite Israel’s universal healthcare system, Palestinians in general, and Palestinian women in particular, experience poorer health compared to other citizens [4]. A previous analysis by the lead author, based on the present data also documented that the prevalence of intimate partner violence (IPV) is twice as high among Palestinian women compared to Jewish Israeli women [6], potentially exacerbating existing health inequalities. However, the relative contribution of IPV to ethnonational health disparities between Palestinian women and their Jewish counterparts has not yet been examined. This study aims to address this critical gap in knowledge.

IPV and Indigenous women’s health

IPV has been defined as a significant public health concern affecting more than one in three women globally [7]. IPV refers to patterns of violent behavior, most often committed by men against a current or former female partner, that exerts control over, and restricts, the partner’s autonomy. IPV can take the form of verbal, economic, physical, or sexual abuse. Rarely a one-time incident, IPV tends to be ongoing and remains widely underreported [8]. Beyond its impact on health and safety, IPV is a profound violation of women’s fundamental human rights [7, 8]. Indigenous women experience a disproportionately high burden of IPV [9, 10] and homicide [11]. A study has shown that First Nations and Métis women in Canada were approximately twice as likely to experience IPV as non-Indigenous women [12], and findings of a systematics review show IPV is associated with poor mental health in the global population of Indigenous women [10]. Higher vulnerability to IPV among postpartum women in Australia was associated with poorer mental health [13].

Global reports consistently show Indigenous women bear a disproportionate burden of poor health and marginalization [14], especially during reproductive ages [15]. Young Indigenous mothers face significant barriers in accessing healthcare services (HCS), contributing to poorer maternal health outcomes [15]. Indigenous women in Australia have lower life expectancy and higher preterm birth [16], while in Canada, Aboriginal women reported higher rates of suicide attempts and depression compared to non-Aboriginal women [17]. A common experience among many Indigenous peoples globally is the enduring impact of colonization, which has negatively affected health through interconnected economic, social, and cultural factors [18]. Given the intersection of gender inequality and structural racism, the health of Indigenous women is a particular concern [15].

IPV and women’s physical and mental health

Around the globe, IPV has been associated with adverse physical and mental health outcomes among women [19–21] and is considered a leading cause of injury, femicide, and suicidality. IPV has been also associated with higher incidence of cancer, anxiety, and other health conditions, as well as poor self-rated health [19]. A national survey of more than 10,000 women in Spain found significant associations between experiencing sexual and psychological IPV and poor self-reported physical and mental health and anxiety [22]. Women who experience IPV during pregnancy and after birth face adverse health outcomes [23], including low birth weight and preterm birth [24]. A national representative sample of Canadian women showed that IPV around the time of pregnancy was associated with increased likelihood of postpartum depression (PPD) [25]. Another study, conducted in Munich, Germany maternity wards, found that women who experienced IPV were five times more likely to have an unplanned pregnancy and 7.3 times more likely to lose the pregnancy [26].

Women who experience IPV also suffer higher multimorbidity (two or more health conditions), whether physical, mental, or both [23]. However, research on the relationship between IPV and women’s multimorbidity is scarce [23]. As well, women experiencing IPV often use HCS more frequently [27], highlighting the need for tailored HCS and IPV detection [28]. Addressing IPV is also crucial for managing chronic health conditions, especially for Indigenous women with limited access to HCS.

While disparities in IPV prevalence between Indigenous women and non-Indigenous are well documented [10], as well as broader Indigenous–non-Indigenous health inequalities [15], little research links IPV to these disparities. A systematic review identified clear association between IPV and increased adverse mental and physical health outcomes among Indigenous women [10].

Health inequalities among women and IPV in Israel

Generally, the health indicators of Palestinians citizens in Israel are poorer compared to their Jewish counterparts. Palestinians have lower life expectancy, higher age-standardized mortality rates for 20 leading causes of diseases, and suffer more from chronic diseases (e.g., obesity, ischemic heart disease) than Jews; however, they face more barriers in accessing HCS [2]. Palestinian women are triply marginalized: by their minoritized status, exposure to structural racism and discrimination, and their lower status within a patriarchal society [29]. Despite Israel’s universal healthcare coverage, which includes primary, secondary and tertiary care, health inequalities persist among Palestinian Arab minority. Research and official reports by the Israeli Ministry of Health (MoH) highlight persistent disparities in access to, and utilization of, healthcare services, consistently identifying Palestinian women as one of the most disadvantaged ethnic groups. They bear a disproportionately high burden of disease morbidity and mortality [2], yet have lower access to preventive services, such as cancer screening and IPV screening [28, 30]. Palestinian women also tend to give birth at a younger age and experience more pregnancy complications [31]. Regarding mental health, compared to Jewish women, Palestinian women are more likely to experience higher psychological distress and anxiety. For perinatal mental health, one study showed 1.5 times more anxiety among Palestinian women compared to Jewish counterparts [32]. Studies have consistently shown Palestinian women suffer more postpartum depression (PPD) compared to Jewish women [29]. Still, Palestinian women are less likely to use mental health services; one study revealed multiple barriers leading to underuse [33]. Although Palestinian citizens make equal or greater use of primary care services, they use specialist and digital services less often than the Jewish population. A high rate of Palestinian women in Israel women cited their unmet need for discreet provision of telehealth services as a barrier to use of these services [34, 35]. The legal status of the local authority also influenced access barriers to healthcare utilization, and residents in unrecognized villages faced significant physical access challenges [33, 36, 37]. Additional barriers include lower digital health literacy, language and cultural differences, and perceived discrimination [29, 38]. Palestinian Arab women who did not use mental health services reported higher levels of multiple barriers compared to those who reported using these services. Barriers included limited knowledge and low awareness of symptoms and treatment, lack of culturally appropriate HCS, insufficient screening, economic barriers, and poverty [33, 37].

In the first national IPV study in Israel, conducted almost a quarter century ago in the early 2000s, Palestinian Arab women had higher IPV prevalence compared to Jewish women. A study based on analysis of the current data set by the first author including a representative sample of 1401 women in the perinatal-period indicated high prevalence of IPV among Palestinian (67%) compared to Jewish immigrant and non-immigrant women (30 and 27%, respectively) [6].

The current study examines the relative contribution of IPV to health inequalities between Indigenous Palestinian and Jewish women citizens in Israel in the perinatal period. We hypothesize that Palestinian women will have poorer health and higher IPV prevalence; that women who experience IPV will have a higher likelihood of experiencing adverse health outcomes; and that IPV will contribute to health inequalities between Palestinian women and Jewish women beyond and above socioeconomic and demographic factors.

Methods

We used data from our original study study on Family Relations, Violence and Health, which was carried out over six months (Oct 2014–Apr 2015) [6]. The study was approved by the Research Ethics Board of Ben-Gurion University of the Negev and Israel’s MoH, Public Health Division. Details of the sample and sampling methods can be found elsewhere [6]. Briefly, the study was conducted in 63 maternal and child health clinics in five MoH districts with a stratified cluster sample of 436 Palestinian and 965 Jewish women who visited the clinics for child follow-ups and immunizations. The sample represents ethnographically and socioeconomically diverse groups of women in Israel in the perinatal period. Sample size in each district was proportional to the local ethnonational (Palestinian vs. Jewish) composition of perinatal-aged women. Trained interviewers conducted face-to-face interviews (30–45 mins in length) in a private room with eligible women (childbearing age; pregnant or postpartum; six weeks to six months after birth; Arabic or Hebrew speaking) who agreed to participate and signed an informed consent form. Women were interviewed in their native language following WHO guidelines [39]. Response rate for Jewish and Palestinian women was 73 and 76%, respectively. Participants received IPV social support services contact information afterwards, and any woman reporting IPV was referred to an MCH nurse, with referrals to a social worker for severe IPV (with women’s consent).

Study measures

We examined inequalities for ten health conditions: Postpartum depression (PPD) was measured by the Edinburgh Postnatal Depression Scale (EPDS) [40]. The EPDS includes ten questions that evaluate postpartum women’s feelings over the past week. The EPDS questionnaire has been translated and tested for reliability in Arabic [41] and Hebrew [42, 43]. Cronbach’s alpha for the total sample was 0.82: 0.84 for Palestinian women, 0.71 for Jewish women. Answer categories ranged from 0 to 3, with a total score of 0–30. The cutoff for the current study was ≥ 10, which we considered a risk for PPD. The EPDS questionnaire was distributed only among post-partum women in the study (N = 1121). Depressive symptoms were measured by the “Center for Epidemiological Studies Depression Scale” (CES-D), which asks about the occurrence of 10 symptoms in the past week [44]. Cronbach’s alpha for the total sample was 0.78: 0.82 for Palestinian women and 0.77 for Jewish women. Answer categories range from 0 to 4, rarely or none (less than 1 day), some or a little (1–2 days), occasionally or a moderate amount (3–4 days), and most or all times (5–7 days). Positively worded items were reversed, scores were summed across all items (minimum = 0, maximum = 25) then dichotomized into low vs. high depressive symptoms using the score of 0.9 as the cutoff. Anxiety was measured by the “State-Trait Anxiety Inventory” (STAI) [45, 46]. STAI measures the presence of anxiety symptoms but is not a diagnostic clinical tool. It includes 10 questions (e.g., I feel calm, I feel secure, I feel tense). Cronbach’s alpha was 0.76 for Jewish women and 0.74 for Palestinian women. Four answer categories ranging from “never” to “almost always.” Coding for answers to positive questions was reversed, and all answers were summed to a total score that was dichotomized using the median score as a cutoff: low anxiety symptoms < 14 vs. high anxiety symptoms ≥ 14 score. Chronic illness was measured by a direct question: “Has the doctor ever told you that you have any chronic illness?” Two categories: 1. Yes (indicate which one or more), 2. None. Self-rated health (SRH) was measured by a direct question: “How do you assess your health in general?” Answers ranged from 1 to 5: 1. Very good, 2. Good, 3. Medium, 4. Not good, 5. Not good at all or bad [47]. Answers were dichotomized into: 1. good health, including category 1, and poor health, including categories 2–5. Unintended/unplanned pregnancy was measured by a direct question: “Did you want this/latest pregnancy?” We dichotomized answers into two categories:1. Yes (for any abortion) and 2. No (for none). Any abortions was measured by a direct question: “How many of your pregnancies ended in clinical abortion?” We dichotomized answers into two categories:1. Yes (for any abortion) and 2. No (for none). Any miscarriages was measured by a direct question: “How many of your pregnancies ended in natural abortion?” We dichotomized answers into two categories: 1. Yes (for any miscarriages) and 2. No (for none). Preterm birth was measured by a direct question: “How many of your pregnancies ended prematurely, with the baby born before week 37 of pregnancy?”. We dichotomized answers into two categories: 1. Yes (for any preterm birth) and 2. No (for none). Multimorbidity was assessed by the sum score of all health issues and dichotomized as follows: women with more than two health issues scored 1; women with one or less health issues scored 0. The main independent variable of ethnonational identity was self-determined, with two categories: Jewish and Palestinian.

We measured IPV using 10 questions about acts ever perpetrated by a participant’s intimate partner. Questions are based on the Preventive Services Task Force Family screening tool [48], which the first author employed in a previous analysis based on the current data on IPV among Palestinian and Jewish women in Israel [6]. We considered a positive answer (yes) to any of the 10 following questions about acts of violence by the intimate partner to be IPV: 1. Are you fearful of drastic changes in your partner’s mood? 2. Does your partner blame you or your environment for his problems? 3. Does your partner try to isolate you from your family and friends? 4. Do you need the permission of your partner for all daily expenses? 5. Is your partner jealous in an extreme manner, to the point that he behaves obsessively, for example, he follows you, calls you frequently, needs to know where you are at all times? 6. Your partner has hit you, kicked you, pushed you or thrown things at you? 7. Your partner threatened to intensify the violence against you if you tell anyone? 8. Your partner threatened that he will commit suicide or hurt himself if you leave him? 9. You partner forced you to have sex with him against your will? 10. You live with a constant sense of danger? More information about the study and tools can be found in our previous publication [6].

Socio-demographic variables included: age, categorized into three groups: 16–24, 25–34, and 35–48 years old; current pregnancy status, answer categories: 1. Pregnant and 2. After birth; and number of children, answers categorized into: 1. 0–1 children, 2. 2–3 children, 3. 4–12 children). Socioeconomic variables included: education, answers divided into three categories: 1. academic degree or university, 2. college or high post-secondary education, and 3. high school or less; current employment, a yes/no question on family income source, with answer categories: 1. Work, 2. Social allowances, or 3. Other source.

Data analysis

Missing cases were less than 10%. First, we compared socioeconomic–demographic characteristics of women by ethnonational identity and any IPV (Table 1). Next, we used univariate analysis to compare health outcomes by ethnonational identity (Palestinian vs Jewish women) (Fig. 1) and examine associations between IPV and each health outcome (Fig. 2). We examined associations between socioeconomic and demographic variables and each health outcome for the total sample, and separately for the two ethnonational groups (Appendix A). Then, using logistic regressions, we calculated unadjusted and adjusted odds ratios (ORs and AORs) and 95% confidence intervals (CIs) for ethnonational (Palestinian vs. Jewish women’s) health inequalities for each of the health outcomes in different models (Table 2): Model 1 was unadjusted, Model 2 controlled for socioeconomic and demographic variables, and Model 3 considered IPV in addition to controlling for the variables in Model 2. Finally, to calculate IPVs relative contribution to the magnitude of health inequalities between Palestinian and Jewish women for each health outcome, we first calculated the reduction (%) of OR from Model 1 to Model 2, and from Model 1 to Model 3, then subtracted the reduction (%) in Model 2 from that in Model 3. The rate difference between the reduction of the AOR in the two models (2 and 3) reveals IPVs net relative contribution beyond socioeconomic and demographic factors (Table 2). Before conducting the regression analysis, we examined correlations between the study’s independent variables; correlation coefficients were lower than our threshold of 0.6, and we included all independent variables in the multivariable analysis, as multicollinearity was unlikely. To assess associations between IPV and health outcomes in the total sample and each group of women, we estimated adjusted and unadjusted ORs and 95%CIs using logistic regressions. Model 1 considered only socioeconomic and demographic variables. Model 2 controlled for IPV in addition to variables in Model 1 (Table 3). The level of significance was set at 5% for all calculations.

Table 1.

Women’s sociodemographic and socioeconomic characteristics in the total sample and among Palestinian and Jewish participants

	Total N = 1041 (100%)	Palestinian women N = 436 (31.1%)	Jewish women N = 965 (68.9%)
Variables	N (%)	N (%)	N (%)	P
Age				< 0.001
16–24	247 (17.6)	163 (37.5)	84 (8.7)
25–34	844 (60.3)	220 (50.6)	624 (64.7)
35–48*	309 (22.1)	52 (12.0)	257 (83.2)
Education				< 0.001
BA/MA/PhD	613 (43.8)	95 (21.8)	518 (53.7)
Postsecondary	251 (17.9)	63 (14.4)	188 (19.5)
High school and less	537 (38.3)	278 (63.8)	259 (26.8)
Family income source				< 0.001
Work	982 (70.1)	340 (78.0)	642 (66.5)
Social allowances	79 (5.6)	53 (12.2)	26 (2.7)
Other	340 (24.3)	43 (9.9)	297 (30.8)
Current employment				< 0.001
Yes	785 (56.7)	110 (25.3)	675 (71.0)
No	600 (43.3)	324 (74.7)	276 (29.0)
Number of children				< 0.001
0–1	560 (40.0)	190 (43.6)	370 (38.4)
2–3	648 (46.3)	168 (38.5)	480 (49.8)
4–12	191 (13.7)	78 (17.8)	113 (11.7)
Current pregnancy status				< 0.001
After birth	1128 (80.8)	275 (63.5)	853 (88.6)
Pregnant*	268 (19.2)	158 (36.5)	110 (11.4)
Intimate partner violence				< 0.001
Yes	558 (39.8)	291 (66.7)	267 (27.7)
No	843 (60.2)	145 (33.3)	698 (72.3)

*Due to missing cases, the total number for the categories of these variables is less than the total number of participants

Fig. 1.

Inequalities in health outcomes among Palestinian and Jewish women in the study sample (N = 1041)

Fig. 2.

Health outcomes among women experiencing any intimate partner violence (IPV) versus not experiencing any IPV in the study total sample of women (N = 1041)

Table 2.

Crude and adjusted odds ratios for ethnonational health inequalities and the relative contribution of IPV

	Postpartum depression § N = 1113	Depression N = 1376	Anxiety N = 1369	Self-Rated Health N = 1370	Unplanned pregnancy N = 1364	Abortions N = 1371	Miscarriages N = 1372	Pre-term birth N = 1372	Chronic illness N = 1290	Multimorbidity N = 1377
	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)	OR (95% CI)
	Model 1‡
Ethnicity
Palestinian	3.57 (2.41–5.31) **	3.46 (2.40–4.99) **	2.06 (1.63–2.60) **	1.16 (0.91–1.47)	4.83 (3.18–7.34) **	1.15 (0.82–1.61)	1.89 (1.47–2.71) **	1.97 (1.51–2.57) **	0.11 (0.05–0.23) **	1.48 (1.10–1.99) *
Jewish	1	1	1	1	1	1	1	1	1	1
	Model 2†
Ethnicity
Palestinian	2.93 (1.80–4.77)**	2.04 (1.27–3.28) *	1.97 (1.47–2.64) **	1.06 (078–1.44)	4.35 (2.51–7.51) **	0.60 (0.37–0.98) *	1.33 (0.96–1.84)	1.61 (1.13–2.29) *	0.12 (0.05–0.28) **	1.40 (0.97–2.01)
Jewish	1	1	1	1	1	1	1	1	1	1
% AORs reduction Model 2 vs 1	17.9%	41.04%	4.36%	8.62%	9.94%	47.8%	29.63%	18.27%	0.09%	5.4%
	Model 3||
Ethnicity
Palestinian	2.21 (1.33–3.66)*	1.46 (0.88–2.40)	1.55 (1.14–2.10)*	0.89 (0.65–1.23)	3.17 (1.79–5.60) **	0.57 (0.38–0.96) *	1.27 (0.91–1.78)	1.51 (1.05–2.19) *	0.10 (0.04–0.23) **	1.04 (0.71–1.53)
Jewish	1	1	1	1	1	1	1	1	1	1
% AORs reduction Model 3 vs 1	38.09%	57.80%	24.75%	23.27%	34.37%	50.43%	32.80%	22.36%	0.09%	29.72%
Δ AORs	20.19%	16.76%	20.39%	14.65%	24.43%	2.6%	3.17%	4.09%	0	24.33%

* p < 0.05. ** p < 0.001. ‡Model 1 Crude. †Model 2 Adjusted for ethnicity, age, education, family income, current employment, current pregnancy status, and number of children. ‖ Model 3 Adjusted for ethnicity, age, education, family income, current employment, current pregnancy status, number of children, and IPV. § For the association between ethnicity and postpartum depression, only women after birth were selected. Δ Calculated for the reduction in AOR between Models 2 and 3.

Table 3.

Unadjusted and adjusted odds ratios for the association between IPV and health outcomes

	Postpartum depression §	Depression	Anxiety	Self–Rated Health	Unplanned pregnancy	Abortions	Miscarriages	Preterm birth	Chronic Illness	Multimorbidity
	OR (95%CI)	OR (95%CI)	OR (95%CI)	OR (95%CI)	OR (95%CI)	OR (95%CI)	OR (95%CI)	OR (95%CI)	OR (95%CI)	OR (95%CI)
	3.a Total sample (N = 1041)
	N = 1113	N = 1376	N = 1369	N = 1370	N = 1364	N = 1371	N = 1372	N = 1372	N = 1290	N = 1377
Model 1– Unadjusted
Any IPV
Yes	3.29 (2.21–4.91)**	3.87 (2.63–5.69)**	2.44 (1.95–0.04)**	1.66 (1.32–2.09)**	4.08 (2.64–6.30)**	1.28 (0.92–1.76)	1.43 (1.13–1.82)*	1.53 (1.18–1.98)**	1.13 (0.79–1.61)	2.52 (1.87–3.39)**
No	1	1	1	1	1	1	1	1	1	1
Model 2–Adjusted‡
Any IPV
Yes	2.36 (1.55–3.62)**	2.64 (1.73–4.01)**	2.12 (1.67–2.70)**	1.68 (1.31–2.15)**	2.76 (1.70–4.49)**	1.13 (0.77–1.65)	1.13 (0.86–1.48)	1.18 (0.88–1.58)	1.80 (1.24–2.62)*	2.44 (1.77–3.36)**
No	1	1	1	1	1	1	1	1	1	1
% OR reduction	28.2%	31.7%	13.1%	+1.20%	32.3%	11.7%	20.9%	22.8%	+59.2%	3.17%
	3.b Palestinian women (N = 436)
	N = 272	N = 430	N = 429	N = 429	N = 419	N = 428	N = 428	N = 428	N = 368	N = 430
Model 1– Unadjusted
Any IPV
Yes	2.13 (1.11–4.08)*	2.90 (1.54–5.47)**	1.82 (1.21–2.74)*	1.72 (1.11–2.65)*	3.00 (1.52–5.93)*	1.51 (0.81–2.83)	1.31 (0.86–2.00)	1.57 (0.99–2.49)	1.45 (0.27–7.62)	2.48 (1.48–4.14)**
No		1	1	1	1	1	1	1	1
Model 2–Adjusted†
Any IPV
Yes	1.99 (1.02–3.87)*	2.47 (1.29–4.74)*	1.88 (1.22–2.90)*	1.72 (1.09–2.72)*	3.96 (1.88–8.35) **	0.85 (0.37–1.98)	1.17 (0.74–1.83)	1.28 (0.79–2.09)	1.41 (0.24–8.15)	2.22 (1.28–3.82)*
No	1	1	1	1	1	1	1	1	1
%OR reduction	6.5%	14.8%	+3.19%	0	+32%	43.7%	10.6%	18.4%	2.75%	10.4%
	3.c Jewish women (N = 965)
	N = 841	N = 946	N = 940	N = 941	N = 945	N = 943	N = 944	N = 944	N = 922	N = 947
Model 1– Unadjusted
Any IPV
Yes	2.80 (1.63–4.80)**	2.80 (1.62–4.83)**	2.25 (1.68–3.01)**	1.68 (1.25–2.26)**	2.30 (1.18–4.46)*	1.15 (0.74–1.76)	1.09 (0.78–1.52)	1.05 (0.73–1.53)	1.82 (1.25–2.66)*	2.46 (1.65–3.66)**
No	1	1	1	1	1	1	1	1	1	1
Model 2–Adjusted†
Any IPV
Yes	2.65 (1.54–4.57)**	2.69 (1.55–4.67)**	2.30 (1.71–3.08)**	1.66 (1.23–2.24)**	2.15 (1.09–4.23)*	1.14 (0.73–1.79)	1.12 (0.80–1.58)	1.06 (0.72–1.56)	1.82 (1.24–2.67)*	2.48 (1.66–3.71)**
No	1	1	1	1	1	1	1	1	1	1
% OR reduction	5.3%	3.9%	+2.2%	1.19%	6.5%	0.8%	+2.75%	+0.9%	0	+0.81%

* p < 0.05. ** p < 0.001

‡ Model 2, for the total sample, was adjusted for ethnicity, age, education, family income, current employment, current pregnancy status, and number of children

† Model 2, for Palestinian and Jewish women, was adjusted for age, education, family income, current employment, current pregnancy status, and number of children

§ Regarding the association between IPV and postpartum depression, only women after birth were selected

% OR reduction was calculated for the difference between unadjusted and adjusted models

Results

Table 1 presents demographic characteristics and socioeconomic status for the Palestinian and Jewish women. We found significant differences between study groups: on average, Palestinian women were younger (mean ±SD = 26.84 ± 5 × 29 vs. 31.11 ± 4.98), had less university-level education, and were less often employed. The primary source of family income in both groups was from work only. More Palestinian than Jewish women reported social allowances as main income source. Palestinian women had more children and were more often pregnant at the interview. Table 1 shows that more than two thirds of Palestinian women reported any IPV, compared to less than one third of Jewish women.

Fig. 1 shows health inequalities between Palestinian and Jewish women. The burden of adverse health outcomes is significantly higher among Palestinian compared to Jewish women. Palestinians reported significantly higher rates of PPD, depressive symptoms, anxiety, unintended pregnancy, any miscarriages, any preterm birth, and multimorbidity.

Jewish women reported higher chronic illness. SRH and abortions did not differ significantly between groups (Fig. 1).

Figure 2 shows women who experienced any IPV had consistent adverse health outcomes compared to women not reporting IPV, including higher PPD, depressive symptoms, anxiety, unintended pregnancy, any miscarriages, preterm birth, SRH, and multimorbidity. No significant difference in chronic illness or any abortions emerged related to any IPV vs. no IPV (Fig. 2).

Table 2 shows unadjusted and adjusted ORs (95%CIs) for health inequalities between Palestinian and Jewish women in different models. Model 1 shows crude ORs for health inequalities between study groups: compared to Jewish women, Palestinian women had significantly higher PPD, depressive symptoms, anxiety, unplanned pregnancy, any miscarriages, preterm birth, and multimorbidity. Chronic illness was significantly higher in Jewish women; no significant differences in any abortions or SRH. Model 2 shows AORs for ethnonational health inequalities between Palestinian and Jewish women considering demographic–socioeconomic factors. Notably, most AORs in Model 2 were attenuated but remained significant, except miscarriages and multimorbidity. Compared to Model 1, the magnitude of the associations (AORs) for health inequalities between Palestinian and Jewish women was attenuated by 17.9% for PPD; 41.04% for depressive symptoms; 4.36% for anxiety; and 9.94% for unplanned pregnancy. For any abortions, the association in Model 2 compared to Model 1 became significant but was reversed: Palestinian women had less abortions than Jewish women (AOR = 0.60, 95% CI = 0.37–0.98). For any miscarriages, the magnitude of the association (AOR) in Model 2 was reduced by 29.63%, and inequalities between groups became non-significant. For preterm birth, the AOR was reduced by 18.27%, and for SHR, 8.62%, but was a non-significant association. We noticed a small change in chronic illness (0.09%). For multimorbidity, inequalities became non-significant after attenuation of the AOR by almost 5.4%.

Model 3 shows further reductions in the magnitude (AORs) of health inequalities after considering IPV in addition to controls (socioeconomic–demographic factors). Compared to Model 1, AORs for inequalities in Model 3 were attenuated 38.09% for PPD, 57.80% for depressive symptoms, 24.75% for anxiety, 34.37% for unplanned pregnancy, 50.43% for abortions, 32.80% for miscarriages (became a non-significant association), 22.36% for preterm birth, 0.09% for chronic illness, 23.27% for SRH (non-significant association), and 29.72% for multimorbidity (non-significant association). To calculate the relative contribution of IPV to ethnonational health inequalities, we subtracted the changes in AORs in Model 3 from those in Model 1 (Δ). This rate difference shows that the specific contribution of IPV was higher for mental health inequalities (20.19% in PPD, 16.76% in depressive symptoms, 20.39% in anxiety, 14.65% in SRH) than physical health inequalities (0 in chronic illness, 2.6% in abortions, 3.17% in miscarriages, 4.09% in preterm birth, 24.43% in unplanned pregnancy). For multimorbidity, the rate difference was 24.33%.

Table 3 shows unadjusted (Model 1) and adjusted (Model 2) ORs (95%CI) for health outcomes among women who reported experiencing any IPV compared to those who did not in the total sample, and separately by study group. For the total sample (Table 3.a), Model 1 shows IPV was significantly associated with adverse health outcomes, except abortions and chronic illness, where associations were not significant. After considering socioeconomic–demographic factors in Model 2, most AORs for health inequalities between women experiencing vs not experiencing any IPV were attenuated but remained significant except miscarriages and preterm births, where the associations became non-significant, and abortions, which remained non-significant.

In Model 2, AORs for health inequalities related to IPV were reduced by 28.2% for PPD; 31.7% for depressive symptoms; 13.1% for anxiety; 32.3% for unplanned pregnancy, and 3.17% for multimorbidity. However, AORs for associations with chronic illness increased by 59.2%, and 1.2% for SRH.

For Palestinian women (Table 3.b), Model 1 shows a significant association between IPV and poorer health outcomes, except for abortions, miscarriages, preterm birth, and chronic illness. After considering contextual factors, AORs for health inequalities were attenuated but remained significant, except for health outcomes that were not significantly associated with IPV in Model 1. The AOR was reduced by 6.5% for PPD, 14.8% for depressive symptoms, and 10.4% for multimorbidity. Notably, the AORs increased by 3.19% for anxiety and 32% for unplanned pregnancy, but remained unchanged for SRH. AORs’ reductions in IPV associations with abortions, miscarriages, preterm birth, and chronic illness were non-significant.

For Jewish women (Table 3.c), Model 1 shows a significant association between IPV and poor health outcomes, except abortions, miscarriages, and preterm birth. After considering socioeconomic–demographic factors, most AORs for health inequalities were attenuated but remained significant, except those not significantly associated with IPV in Model 1. The AOR was reduced by 5.3% for PPD; 3.9% for depressive symptoms, 6.5% for planned pregnancy, and 1.19% for SRH. The AOR increased 2.22% for anxiety and 0.81% for multimorbidity, while remaining unchanged for chronic illness. Small reductions in the associations with abortions, miscarriages, and preterm birth were non-significant.

Discussion

This study is the first we know of to assess the relative contribution of intimate partner violence (IPV) to health inequalities between Palestinian and Jewish women. By calculating the percentage of change in adjusted odds ratios (AORs) for health inequalities between adjusted models (accounting for socioeconomic–demographic factors) and models that adjusted for IPV in addition to these factors, we assessed the unique impact of IPV on health inequalities between Palestinian and Jewish women across ten health conditions, including multimorbidity. We found IPV prevalence was significantly higher among Palestinian, compared to Jewish, women, aligning with previous studies in Israel [6, 49]. While this reflects broader evidence related to higher IPV among Indigenous women [10], the novelty of our study lies in its focus on health inequalities between two ethnonational groups of women and its assessment of IPVs relative contribution to these disparities. A primary finding is that IPV plays a significant role in the observed health disparities between Palestinian and Jewish women. IPV contributed to nearly 24.4% of inequalities in multimorbidity. While IPVs impact on physical health inequalities ranged from 0 to 24.4%, its contribution to mental health disparities was more pronounced, ranging from 14.65–20.39%. These findings highlight IPVs central role in shaping health disparities between Palestinian and Jewish women, likely influenced by the intersection of gender inequities and structural racism [29]. Although our findings are specific to the context of Palestinian and Jewish women, they underscore the importance of investigating similar dynamics in other populations, particularly given the extensive evidence that Indigenous and minority women experience poorer health outcomes than non-Indigenous counterparts.

Our findings also reveal higher prevalence of adverse health outcomes- including multimorbidity, among women who have experienced IPV. These associations were particularly pronounced in mental health outcomes, obstetric outcomes, and multimorbidity. While our findings are consistent with many previous research findings on the relationship between IPV and adverse health [19, 20], they also beg the question: how does IPV contribute more to the excess of health inequalities in Palestinian women? One explanation would be that Palestinian women not only experience poorer health outcomes, but also face multiple marginalizations, including systemic racism, economic disadvantage, and political and social marginalization as women, that limit their ability to protect themselves from adverse IPV effects and sustain good health [29]. Previous research has shown Palestinian women lack social support compared to Jewish women [6], for example, which might be critical in their inability to mitigate IPV disparities, as social support can be a protective factor against various IPV forms [50]. Work and family dynamics can also contribute to these disparities. Palestinian women generally have lower labor-force participation, lower educational attainment, larger families, and higher pregnancy rates, all of which likely influenced the reported health outcomes and inequalities. Beyond these socio-cultural factors, barriers to social and HCS access further shape the impacts of IPV on Palestinians women’s health and health inequalities [33]. Factors such as trust in HCS, accessibility of services [51], institutional discrimination [32, 52], and availability of IPV-related support resources all play a crucial role in shaping health outcomes in Indigenous women [10, 15]. Jewish women may be more aware of their health rights and symptoms, while Palestinian women experiencing IPV are often reluctant to report it or seek help [53, 54]. This reluctance relates to language barriers, as many IPV-related services in Israel are primarily Hebrew language–based and not tailored to Arab society needs [27]. Reports from the Research and Information Center of the Knesset have highlighted the dearth of Arabic-language HCS and information [55]. By contrast, Jewish women experience fewer structural constraints in seeking help.

Worldwide, IPV has been consistently associated with adverse mental health outcomes, particularly among women from Indigenous and marginalized populations [18, 20]. A recent meta-analysis, found that women exposed to IPV had 1.97 times higher odds of experiencing depression, and between 2.15 and 2.66 times higher odds of PTSD symptoms compared to those not exposed. The review further noted that women from Indigenous backgrounds are disproportionately affected by IPV, mainly due to intersecting social disadvantages, such as poverty, discrimination, and barriers to accessing services [20]. In the Israeli context, evidence shows Palestinian women experience disproportionately high rates of IPV compared to Jewish women [54]. These disparities were not explained by individual characteristics alone, but were associated with neighborhood-level disadvantage, low social cohesion, and broader processes of ethnic discrimination and marginalization [54]. These findings emphasize that IPV is not an isolated or individual-level issue, but one that is embedded in, and shaped by, systemic and historical forms of marginalization. In such contexts, IPV serves both as a consequence of broader social conditions and a mechanism that deepens health disparities, particularly among ethnically minoritized groups.

Our findings showing that IPV contributes more to mental health disparities than physical health inequities are particularly concerning, as untreated mental health conditions can reinforce cycles of poverty and poor health that intensify health inequalities. Previous research found that physical IPV is associated with increased adverse mental and physical health outcomes [20], suggesting that distinguishing between types of IPV might be essential for understanding health inequalities related to specific outcomes. While our study examined any IPV, without differentiating between types, the lead author’s previous analysis based on the present data indicates that Palestinian women experience higher rates of all IPV forms (physical, psychological, social) [6]. Future research should explore specific impacts of different IPV types on health disparities among women in Israel and how each contributed to the excess to health inequalities.

Interestingly, in our study, Jewish women exhibited higher prevalence of chronic illness than Palestinian women (15% vs. 1.9%), and the associations between IPV and poor health were stronger among Jewish women. Several factors might explain this. First, age-related differences in chronic illness prevalence might relate to the fact that the average age of Jewish women in our sample was higher than that of Palestinian women (M, SD = 31.1, 4.98 vs. 26.8, 5.29, respectively), correlating with an increased likelihood of chronic illness. Second, Palestinian women might have under-reported chronic illness, which, in turn, might relate to stigma and healthcare access barriers, as low-income minority women are less likely to disclose chronic illnesses [56].

Finally, inclusion of multimorbidity as a health outcome strengthens our findings. Multimorbidity captures the complexity of IPVs health effects more comprehensively than single health outcomes [23]. Our results confirm that women who experience IPV have a significantly higher risk of multiple health conditions, reinforcing the urgent need to address IPV as a public health priority.

Conclusion

Our findings position IPV not merely as an individual or interpersonal issue, but as a structural and overlapping social determinant of health that reinforces existing inequities between Palestinian and Jewish women in Israel. IPV interacts with, and is amplified by, other systemic determinants, such as poverty, discrimination, and restricted access to healthcare, that collectively produce compounded health disadvantages for minoritized women. Tackling IPV therefore requires a comprehensive approach that addresses the broader social conditions and power structures that underpin both violence and health disparities. Advancing health equity demands that future policies and interventions move beyond individual-level solutions to include systemic changes that dismantle the root causes of IPV and structural inequality.

Abbreviations

IPV

Intimate partner violence

HCS

Healthcare services

MoH

Ministry of Health

AOR

Adjusted odds ratio

Author contributions

ND was the PI on the study, she designed the study, led the data analysis and prepared the first draft of the paper. She reviewed the different versions and prepared the final version of the paper. BA and SEZ helped with conducting the data analysis. BA, NAS, HAL, and SEZ helped in writing different sections of the paper. All authors contributed to conceptualization of the manuscript and approved the final version.

Funding

This study was funded by Israel Science Foundation Grant 881/13.

Data availability

All data generated or analysed during this study are included in this published article [and its supplementary information files].

Declarations

Ethics approval and consent to participate

The study was approved by the Ethics Board of Ben-Gurion University of the Negev and the Public Health Division of the Israeli Ministry of Health. All participants signed a consent form before that participated in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.