Psychological Aspects of the Israeli–Palestinian Conflict: A Systematic Review

Eligibility Criteria and Study Selection

Search Strategy

We conducted our systematic searches in the PubMed database, using the following search terms (and their plural equivalents) from three categories:

Geographical region: Israel, Palestine, Palestinian, West Bank, and Gaza

The full list of psychological domain search terms was determined through an iterative process. We began by searching for broad psychological terms such as “distress” and “mental health” as well as terms we expected to be relevant based on previous literature (e.g., “moral injury” and “anxiety”). We then reviewed key words from these articles and relevant studies identified through a scan of the reference section (see subsequently) to augment our list of search terms. We did this until we reached a point of saturation, where adding search terms did not yield any new articles. We restricted the final search to those articles with the aforementioned search terms in their title or abstract. To augment our PubMed search, we scanned the reference sections of each identified article and reviewed studies that appeared to be relevant based on their titles. As noted earlier, we limited our search to research published in or after 1990. In addition, we excluded nonempirical studies (e.g., theoretical papers or reviews) and studies that did not examine the association of psychological dimensions with the Israeli–Palestinian conflict.

Data Items

The following data were sought within each reviewed study: year(s) of data collection, sample size, ethnicity/nationality of participants (e.g., Palestinian, Arab Israeli, and Jewish Israeli), age range, gender, study design (e.g., longitudinal and cross-sectional), psychological dimensions measured, findings on the relationship between conflict/violence and psychological dimensions, and findings on gender, age, or ethnic differences in psychological dimensions. After a study was reviewed, it was coded according to the type of psychological dimensions it measured: psychiatric disorders (e.g., PTSD and depression), functional impairment (e.g., family functioning and general resource loss), general psychological dimensions (e.g., distress, shame, hopelessness, and psychological resource loss), behaviors (e.g., risk taking and violence/aggression), and identity and attitudes (e.g., ethnocentrism and attitudes toward peace).

Data Collection Process

Following the search described earlier, the first author reviewed the abstracts of studies in the search results and excluded those that were clearly not relevant. Next, three research assistants obtained the full-text articles, independently reviewed them, and documented the data items described earlier in a spreadsheet. For quality assurance, the first author also reviewed all articles. In cases of ambiguity, uncertainty, or disagreement about a data item or the inclusion of a study in the review, the issue was discussed until consensus was reached.

Six hundred twenty-five articles were initially screened and assessed for eligibility (see Figure 1 for a diagram of the article selection process). This set of articles was narrowed to 382 after excluding those that were clearly ineligible upon a review of their titles and abstracts (e.g., nonempirical papers, those not investigating psychological dimensions in the relevant geographical region.) Of these, 326 were excluded after a more thorough review of the full-text articles revealed their ineligibility based on our selection criteria. Ultimately, 56 studies were included in the review.

OPEN IN VIEWER

Figure 1.

Literature review flow diagram.

Psychiatric Disorders

Twenty-nine (52%) of the studies we reviewed examined the association between degree of exposure to the regional conflict and violence and one or more psychiatric disorders in Israelis and/or Palestinians (see those denoted “PD” in Tables 1–3). Twenty-five of these studies used a cross-sectional design, four used a longitudinal design, and none used a repeated cross-sectional design. The most common disorders examined were PTSD (26 studies) and depression (13 studies). The following disorders were examined in one study only: separation anxiety (Berger et al., 2007), conduct disorder (Miller, el-Masri, Allodi, & Qouta, 1999), attention-deficit/hyperactivity disorder (Miller et al., 1999), acute stress reaction (Musallam et al., 2005), and somatoform disorders (Punamaki et al., 2008). The majority (n = 22, 76%) of the 29 studies that examined psychiatric disorders, including all 4 longitudinal studies, found that exposure to political violence was associated with elevated psychiatric symptoms (Bleich, Gelkopf, Berger, & Solomon, 2008; Canetti-Nisim et al., 2009; Cohen & Eid, 2007; Gelkopf, Berger, Bleich, & Silver, 2012; Hamama-Raz, Solomon, Cohen, & Laufer, 2008; Henrich & Shahar, 2013; Hobfoll, Canetti-Nisim, & Johnson, 2006; Khamis, 2005; I. Lavi, Canetti, Sharvit, Bar-Tal, & Hobfoll, 2014; T. Lavi & Solomon, 2005; Miller et al., 1999; Musallam et al., 2005; Pat-Horenczyk, Abramovitz, et al., 2007; Punamaki et al., 2008; Qouta et al., 2007; Schiff, 2006; Schiff et al., 2006; Slobodin, Caspi, Klein, Berger, & Hobfoll, 2011; Solomon & Lavi, 2005; Thabet, Abed, & Vostanis, 2002; Thabet & Vostanis, 1999, 2000). Only seven of the studies (all cross-sectional) found no association between psychiatric disorders and exposure to political violence (Berger et al., 2007; Bleich et al., 2006; Bleich, Gelkopf, & Solomon, 2003; Elbedour, Onwuegbuzie, Ghannam, Whitcome, & Abu Hein, 2007; Qouta et al., 2003; Shalev, Tuval, Frenkiel-Fishman, Hadar, & Eth, 2006; Somer et al., 2007).

Eight (28%) of the studies reporting on psychiatric disorders examined ethnic differences. All of these studies were cross-sectional. Five of the eight studies found that Palestinians and Arab Israelis were at greater risk for mental health problems compared to Jewish Israelis (Bleich et al., 2006; Hamama-Raz et al., 2008; Hobfoll et al., 2006; I. Lavi et al., 2014; Somer et al., 2007). Two of the eight studies found similar rates of psychiatric disorders among different ethnic groups (Bleich et al., 2003; Cohen & Eid, 2007) and one study found that Palestinians were at higher risk for PTSD, but not depression, compared to Israeli Palestinians (T. Lavi & Solomon, 2005).

Functional Impairment

“Functional impairment” refers to a person’s inability to carry out functions in important areas of life, such as work, education, or relationships. Eight (14%) of the studies we reviewed tested the association between exposure to regional political conflict and violence and functional impairment. All of these studies used a cross-sectional design (see those denoted “F” in Table 1). Two studies focused specifically on problems in family functioning (Al-Krenawi, Graham, & Sehwail, 2007; Punamaki, 1996), two studies examined resource loss (Heath, Hall, Russ, Canetti, & Hobfoll, 2012; Slobodin et al., 2011), one study focused on disruptions in daily living (Shalev et al., 2006), and three studies reported on overall functional impairment (Bleich et al., 2008; Gelkopf et al., 2012; Pat-Horenczyk, Abramovitz, et al., 2007). One of these three studies also reported on health care utilization (Gelkopf et al., 2012), which for the purposes of this review we include as a measure of functional impairment. Nearly all the studies in this group (n = 7, 88%) found that greater exposure to regional political conflict and violence was associated with more severe functional impairment (Al-Krenawi et al., 2007; Bleich et al., 2006; Gelkopf et al., 2012; Heath et al., 2012; Pat-Horenczyk, Abramovitz, et al., 2007; Shalev et al., 2006; Slobodin et al., 2011). One study found no relationship between exposure to regional political conflict and violence and problems with family functioning (Punamaki, 1996). None of these eight studies sampled multiple ethnic groups, and thus ethnic differences in functional impairment were not investigated.

General Psychological Dimensions

Forty-four (79%) of the studies we reviewed estimated the relationship between regional political conflict and violence and more general psychological dimensions, including both positive (e.g., resilience) and negative (e.g., anxiety) dimensions (see those denoted “GP” in Tables 1–3). Thirty-four of these studies used a cross-sectional design, eight used a longitudinal design, and two used a repeated cross-sectional design. Fifteen studies examined “psychological symptoms” or “mental health symptoms” (Al-Krenawi et al., 2007; Berger et al., 2007; Giacaman, Shannon, Saab, Arya, & Boyce, 2007; T. Lavi & Solomon, 2005; Miller et al., 1999; Musallam et al., 2005; Pat-Horenczyk, Abramovitz, et al., 2007; Punamaki, 1990a, 1990b, 1996; Punamaki, Komproe, Qouta, El-Masri, & de Jong, 2005; Slone, Shoshani, & Lobel, 2013; Slone, Shoshani, & Paltieli, 2009; Soffer-Dudek & Shahar, 2010; Thabet & Vostanis, 1999), thirteen studied psychological distress (Braun-Lewensohn & Sagy, 2011; Gelkopf et al., 2012; Giacaman, Husseini, Gordon, & Awartani, 2004; Hammoudeh, Hogan, & Giacaman, 2013; Heath et al., 2012; Khamis, 1998; Punamaki, 1990a, 1990b; Punamaki et al., 2008; Sagy & Braun-Lewensohn, 2009; Shalev et al., 2006; Shrira, Shmotkin, & Litwin, 2012; Yagur, Grinshpoon, & Ponizovsky, 2002), and nine studied anxiety (Ben-Zur & Gilbar, 2011; Braun-Lewensohn & Sagy, 2011; Elbedour et al., 2007; Henrich & Shahar, 2013; T. Lavi & Solomon, 2005; Punamaki, 1996; Sagy & Braun-Lewensohn, 2009; Slobodin et al., 2011; Thabet et al., 2002). Numerous other psychological dimensions were examined (e.g., future orientation, resilience, sense of safety, worries, social support, etc.), though none of these are taken up by more than one or two studies. Most studies found that greater exposure to regional political conflict and violence was significantly related to poorer psychological health. For example, of the 13 studies on psychological distress, 10 found a significantly positive association with exposure to regional political conflict and violence (Gelkopf et al., 2012; Giacaman et al., 2004; Hammoudeh et al., 2013; Heath et al., 2012; Khamis, 1998; Punamaki, 1990b; Punamaki et al., 2008; Sagy & Braun-Lewensohn, 2009; Shrira et al., 2012; Yagur et al., 2002), 2 found no association (Braun-Lewensohn & Sagy, 2011; Punamaki, 1990a), and 1 found a negative association (Shalev et al., 2006). Among the nine studies that examined anxiety, seven found a positive association with exposure to regional political conflict and violence (Ben-Zur & Gilbar, 2011; Elbedour et al., 2007; Henrich & Shahar, 2013; T. Lavi & Solomon, 2005; Punamaki, 1996; Sagy & Braun-Lewensohn, 2009; Slobodin et al., 2011), one found a negative association (Thabet et al., 2002), and one found no association (Braun-Lewensohn & Sagy, 2011). Only 7 of the 44 studies that examined general psychological dimensions reported on ethnic differences (5 cross-sectional, 2 longitudinal, and 0 repeated cross-sectional). Two of these found that Palestinians exposed to regional conflict and violence were at greater risk for psychological problems than were Israelis (Boxer et al., 2013; Hobfoll et al., 2006), two found that Arab Israelis exposed to regional conflict and violence were at greater risk for psychological problems than were Jewish Israelis (Bleich et al., 2006; Hobfoll et al., 2009), and three reported inconsistent findings, with one group faring worse than the other on one dimension but better (Shrira et al., 2012) or the same (Braun-Lewensohn & Sagy, 2011; T. Lavi & Solomon, 2005) on another dimension.

Behaviors

Twelve (21%) of the reviewed studies investigated the association between exposure to regional conflict and violence and a range of behaviors (see those denoted “B” in Tables 1–3). Eight of these studies used a cross-sectional design, three used a longitudinal design, and one used a repeated cross-sectional design. Coping was the most common behavior measured (Bleich et al., 2003, 2008; Musallam et al., 2005; Punamaki, 1990a, 1990b, 1996; Punamaki & Puhakka, 1997), though some coping behaviors were likely to be more adaptive (e.g., seeking support) than others (e.g., using drugs and alcohol to cope). Findings on coping were mixed. Greater exposure to the regional conflict and violence was associated with increased ignoring thoughts of violent events (Bleich et al., 2008), using drugs and alcohol to cope (Bleich et al., 2008), looking for more information (Bleich et al., 2008), venting (Musallam et al., 2005), seeking emotional support (Musallam et al., 2005), political activity (Punamaki, 1990b), situational coping (Punamaki et al., 2008), social affiliation (Punamaki & Puhakka, 1997), and problem restructuring (Punamaki & Puhakka, 1997). However, exposure to the regional conflict and violence was negatively associated with other coping behaviors such as passivity (Punamaki, 1990b), active fighting (Punamaki & Puhakka, 1997), and hostile confrontation (Punamaki & Puhakka, 1997). Two studies found no association between exposure to the regional conflict and violence and coping (Bleich et al., 2003; Punamaki, 1990a). Other less commonly examined behaviors included aggression/violence (Boxer et al., 2013; Henrich & Shahar, 2013), risk taking (Pat-Horenczyk, Peled, et al., 2007), alcohol use (Schiff, 2006; Schiff et al., 2006), and help seeking (Bleich et al., 2003). There was no association between exposure to the regional conflict and violence and help seeking (Bleich et al., 2003). Both longitudinal studies on aggressive/violent behavior reported that exposure to the regional conflict and violence predicted an increase in aggressive and violent behavior (Boxer et al., 2013; Henrich & Shahar, 2013), though one study on violent behavior also found no relationship between exposure to conflict and violence and aggression (Henrich & Shahar, 2013). Exposure to conflict and violence was positively associated with risk taking (Pat-Horenczyk, Peled, et al., 2007) and alcohol use (Schiff, 2006; Schiff et al., 2006). Only one study looked at ethnic differences in behavior and found that Palestinians exposed to the regional conflict and violence reported higher levels of aggressive behaviors than did Israelis (Boxer et al., 2013).

Attitudes

Nine (16%) of the studies we reviewed examined the relationship between attitudes and exposure to the regional conflict (see those denoted “ATT” in Tables 1–3). Six of these studies used a cross-sectional design, one used a longitudinal design, and one used a repeated cross-sectional design. The attitudes examined in these studies pertained primarily to political processes (e.g., attitudes toward peace or peace talks and support for political violence) or to an individual’s attitude toward his or her own religious, national, or ethnic identity (e.g., ethnocentrism and patriotic involvement) or toward others (e.g., fear/hatred or negative stereotypes about the out-group and ability to forgive). Greater exposure to regional political conflict and violence was associated with an increase in the following attitudinal variables: exonerating cognitions (Roccas, Klar, & Liviatan, 2006), ethnocentrism (Hobfoll et al., 2006), support for political violence (Hobfoll et al., 2006), authoritarianism (Hobfoll et al., 2006), Palestinians’ (but not Israelis’) threat perceptions (I. Lavi et al., 2014), Palestinians’ and Israelis’ fear/hatred toward the out-group (I. Lavi et al., 2014), Palestinians’ and Israeli Arabs’ negative stereotypes about Israeli Jews and Israeli Jews’ negative stereotypes about Palestinians (Niwa et al., 2014), glorification of war (Punamaki, 1996), and pessimism about peace (T. Lavi & Solomon, 2005; Punamaki, 1996; Solomon & Lavi, 2005). Greater exposure to the regional conflict was associated with less forgiveness among Palestinians but not among Jews (Hamama-Raz et al., 2008). No relationship was found between exposure to the regional conflict and identification with one’s national group (Roccas et al., 2006), Jews’ exclusionist political attitudes toward Palestinians (Canetti-Nisim et al., 2009), attitudes toward peace (Hamama-Raz et al., 2008), revenge seeking (Hamama-Raz et al., 2008), and attitudes toward the enemy (Punamaki, 1996). Only four studies examined ethnic differences in attitudes. One found that Palestinians reported being less able to forgive but more willing to accept peace and revenge seeking compared to Jews (Hamama-Raz et al., 2008); a second found greater perceived threat and more fear and hatred toward the out-group among Palestinians than among Israelis (I. Lavi et al., 2014); and a third reported more positive attitudes toward peace among Israeli Palestinians than among non-Israeli Palestinians (T. Lavi & Solomon, 2005). One study found that negative stereotypes were more prevalent among Israeli Jewish youth than among Palestinian youth but that Palestinian youth held more negative stereotypes than did Israeli Arab youth (Niwa et al., 2014). This same study also found that while the negative stereotypes held by Israeli Jewish and Palestinian youth about each other increased over the course of the study, negative stereotypes held by Israeli Arabs about their Jewish peers declined over time.

Summary of Evidence

As a whole, the body of literature we reviewed suggests that increased exposure to the Israeli–Palestinian conflict may have detrimental effects on psychological well-being across all the domains we examined. These findings are not surprising in light of the many other studies from across the globe which show that exposure to violence has negative mental health consequences for children and adults, individuals, and communities (Dubow et al., 2009; Muldoon, 2013; Sousa et al., 2013). However, our review indicates that the effects of exposure to chronic conflict and violence likely extend beyond the psychiatric disorders and symptoms most commonly studied. In the body of work we reviewed, we found evidence that exposure to chronic conflict and violence informs the way Israelis and Palestinians think, feel, and act, including their attitudes toward different religious and ethnic groups and their degree of support for peace or war.

Our review suggests that there is a need for more studies on the nonclinical psychological consequences of the conflict. For instance, while 29 studies included measures of common mental disorders such as PTSD and depression, only 9 assessed individuals’ attitudes. A large proportion of the literature examined a heterogeneous group of what we called “general psychological dimensions,” but even this category consisted primarily of subclinical symptoms (e.g., “psychological distress”), which are ultimately related to known psychiatric disorders. Given the fact that the majority of Israelis and Palestinians will not develop disorders like PTSD from exposure to the conflict (Thabet & Vostanis, 2000), a clearer picture of the broad impact of the Israeli–Palestinian conflict requires more research on individuals’ attitudes, beliefs, feelings, and behaviors. Research on the mechanisms and circumstances by which the conflict has had an impact on Israeli and Palestinian attitudes and behavior toward one another could assist practitioners and policy makers in implementing strategies to mitigate the negative effects of the conflict across the population, not just in those who are vulnerable to developing psychiatric disorders.

To accurately estimate the effect of exposure to the decades-long regional conflict and violence on the psychological well-being of Israelis and Palestinians, longitudinal study designs are needed. Unfortunately, of the 56 studies we reviewed, only 10 employed a longitudinal design. It is difficult to draw firm conclusions about the effect of exposure to chronic conflict on psychological well-being with so few longitudinal studies, given the variability in their measures, study populations, and the timing of their data collection. The start dates for data collection in these studies ranged from 1982 to 2008, a period in which there were spurts of severe political violence and terror interspersed with periods of relative peace. Additional longitudinal studies are needed to better understand the conditions under which—and the types of groups for whom—exposure affects psychological well-being.

The work we reviewed also suggests that Palestinians may be particularly vulnerable group. This finding is consistent with literature suggesting that individuals who experience discrimination or social inequality, or who live in impoverished conditions with a lack of access to resources and services, are at elevated risk for mental health problems (Sanders-Phillips, Settles-Reaves, Walker, & Brownlow, 2009; Schuster et al., 2012). However, this finding should be taken with caution, only 12 of the studies we reviewed included multiple ethnic groups in their samples, and some of these did not report findings on ethnic differences. It is also important to note that comparisons between racial/ethnic groups have been used to justify negative stereotypes and the secondary social status of minorities. These comparisons often fail to take into account confounding factors such as income, experiences of discrimination, and social inequality, which may be much stronger predictors of psychological well-being than race or ethnicity. For this reason, within-group variation can be just as informative as between-group variation. Unfortunately, in this review, we were unable to interpret findings on Palestinians or Israelis across studies, even those examining the same psychological construct, due to a lack of consistency in measurement, a well-documented problem within studies on exposure to violence worldwide (Pinheiro, 2006).

Limitations

Our study had limitations. First, our selection criteria limit the generalizability of our findings. We reviewed only studies published since 1990 that specifically measured the association between exposure to the Israeli–Palestinian conflict and psychological dimensions. We may have missed informative findings from older studies or those that did not quantify the association of interest. Also, we did not examine the size of the associations or effects reported in the literature. A meta-analysis would be a more rigorous method for estimating such effects, though a meta-analytic approach may be difficult in this context due to the heterogeneity in constructs. Finally, due to the dearth of longitudinal studies noted earlier, it is premature to draw conclusions about the effect of exposure to chronic regional conflict and violence on psychological outcomes among Israelis and Palestinians. Our findings primarily reflect cross-sectional associations where directionality cannot be determined.