Moral injury and its mental health consequences among protesters: findings from Israel’s civil protest against the government's judicial reform

ABSTRACT

Background: Government actions and participating in protracted-duration protests against it affect protesters' mental health, leading to high distress levels, such as posttraumatic and depressive symptoms. Aside from exposure to violence and other issues, protest participation can pose unique challenges to the protesters as they may be exposed to potentially morally injurious events (PMIEs), such as the betrayal of leaders they once trusted. This study's primary objective was to examine the extent of psychological difficulties among civilians participating in long-duration protests in Israel. More specifically, the study aimed to understand the contribution of exposure to protest-related PMIEs to psychological difficulties such as posttraumatic and depressive symptoms.

Method: Participants comprised 4036 Israelis who were actively involved in the unfolding civil protest movement against the government-led judicial overhaul between January 2023 and August 2023. The protesters completed validated self-report questionnaires that included measures of PMIE exposure, PTSD and depressive symptoms.

Results: About half (44.3%) of the sample met the criteria for self-report diagnosis of major depression and 10.6% for PTSD. Most of the protesters indicated their exposure to at least one moral injury event, with 63.9% reporting exposure to PMIE-Betrayal. Protesters exposed to PMIEs reported significantly higher levels of both PTSD and depression than non-PMIE-exposed protesters. Hierarchical regression analyses revealed that, beyond demographics and protest-related characteristics such as exposure to violence, PMIE dimensions significantly contributed to both PTSD and depression levels.

Conclusions: The findings highlight the mental burden of protesters during the civil protests against the judicial overhaul in Israel. More central to the present research, the findings highlight the critical contribution of PMIEs exposure to this burden. Clinicians treating protesters coping with depression and PTSD following the civil actions should attend to their exposure to PMIEs, which may relate to the deleterious psychological effects among protesters.

HIGHLIGHTS

• Participation in protests have high mental burden as about half of the sample met the criteria for self-report diagnosis of major depression and 10.6% for PTSD.

• Protesters exposed to protest-related PMIEs reported significantly higher levels of both PTSS and depression than non-PMIE-exposed protesters.

• The findings highlight the critical contribution of PMIE-betrayal exposure to both PTSS and depressive symptoms, above and beyond demographic and protest-related characteristics.

Antecedentes: Las acciones gubernamentales y la participación en protestas prolongadas en su contra afectan la salud mental de los manifestantes, lo que genera altos niveles de angustia, como síntomas postraumáticos y depresivos. Aparte de la exposición a la violencia y otros problemas, la participación en protestas puede plantear desafíos únicos para los manifestantes, ya que pueden estar expuestos a eventos potencialmente perjudiciales para la moral (PMIEs por sus siglas en inglés), como la traición de los líderes en los que alguna vez confiaron. El objetivo principal de este estudio fue examinar el alcance de las dificultades psicológicas entre los civiles que participan en protestas de larga duración en Israel. Más específicamente, el estudio buscó entender la contribución de la exposición a los PMIEs, relacionados con protestas a las dificultades psicológicas como los síntomas postraumáticos y depresivos.

Método: Los participantes fueron 4036 israelíes que participaban activamente en el movimiento de protesta civil contra la reforma judicial liderada por el gobierno entre enero y agosto del 2023. Los manifestantes completaron cuestionarios de autoinforme validados, que incluían medidas de exposición a los PMIEs y síntomas de trastorno de estrés postraumático (STEPT) y síntomas depresivos.

Resultados: Aproximadamente la mitad (44,3%) de la muestra cumplió con los criterios para el diagnóstico autoinformado de depresión mayor y el 10,6% para el trastorno de estrés postraumático. La mayoría de los manifestantes indicaron su exposición a al menos un evento de daño moral, con un 63,9% reportando exposición a PMIEs relacionados con traición. Los manifestantes expuestos a PMIEs informaron niveles significativamente más altos de STEPT y depresión que los manifestantes no expuestos a PMIEs. Los análisis de regresión jerárquica revelaron que, más allá de la demografía y las características relacionadas con las protestas, como la exposición a la violencia, las dimensiones del PMIEs contribuyeron significativamente tanto al STEPT como a los niveles de depresión.

Conclusiones: Los hallazgos resaltan la carga mental de los manifestantes durante las protestas civiles contra la reforma judicial en Israel. Lo más importante para la presente investigación es que los hallazgos resaltan la contribución crítica de la exposición a los PMIEs a esta carga. Los médicos que tratan a manifestantes que enfrentan depresión y trastorno de estrés postraumático después de acciones civiles también deben prestar atención a su exposición a los PMIEs, que pueden estar relacionados con efectos psicológicos nocivos.

背景：政府行为和参与针对政府的长期抗议会影响抗议者的心理健康，导致高度痛苦，例如创伤后和抑郁症状。除了面临暴力和其他问题之外，抗议参与还可能给抗议者带来独特的挑战，因为他们可能会面临潜在的道德伤害事件（PMIE），例如他们曾经信任的领导人的背叛。 本研究主要旨在考查参加以色列长期抗议活动的平民的心理困难程度。更具体地说，该研究旨在了解接触与抗议相关的PMIE对创伤后和抑郁症状等心理困难的影响。

方法：参与者包括 2023 年 1 月 23 日至 8 月 23 日期间积极参与反对政府主导司法改革的民间抗议运动的 4036 名以色列人。抗议者完成了经过验证的自我报告调查问卷，其中包括 PMIE 暴露、PTSD 症状（ PTSS）和抑郁症状的测量。

结果：大约一半（44.3%）的样本符合自我报告重性抑郁诊断标准，10.6% 的样本符合 PTSD。 大多数抗议者表示，他们至少经历过一次道德伤害事件，其中 92.1% 的人报告经历过 他人PMIE，63.9% 的人报告经历过 背叛PMIE。 PMIE 暴露的抗议者报告的 PTSS 和抑郁水平显著高于未接触 PMIE 的抗议者。分层回归分析显示，除了人口统计和抗议相关特征（例如暴力暴露）之外，PMIE 维度对 PTSS 和抑郁水平也有显著影响。

结论：调查结果强调了抗议者在反对以色列司法改革的民间抗议期间所承受的精神负担。对本研究而言更重要的是，这些发现强调了 PMIE 暴露对这一负担的关键贡献。治疗民事诉讼后患有抑郁和创伤后应激障碍 (PTSD) 的抗议者的临床医生也应该注意他们 PMIE暴露情况，这可能与有害的心理影响有关。

KEYWORDS:

• Protest
• moral injury
• violence
• PTSD
• depression

PALABRAS CLAVE:

• Protesta
• daño moral
• violencia
• trastorno de estrés postraumático
• depresión

关键词:

• 抗议
• 道德伤害
• 暴力
• PTSD
• 抑郁

1. Background

Collective actions such as protests and demonstrations have become highly prevalent (Tarrow, 2011) in recent decades, with studies indicating that such protests encompass most of the world’s population (Banks & Wilson, 2020). Aside from changes in societies and nations, such involvement in protests and collective actions can have a deleterious psychological impact on the protesters (e.g. Ojeahere et al., 2021). In a recent systematic review, Ni et al. (2022a) highlight that protesters can suffer from adverse mental health outcomes, such as depression and posttraumatic stress symptoms (PTSS), even in non-violent protests. Given the pervasiveness of protests internationally, this knowledge may underscore the importance of understanding the facilitators of such adverse mental health outcomes among protesters.

Several general and specific risk factors for high levels of depression and PTSS among protesters have been identified. Ni et al.’s (2022a) systematic review of 52 studies from various international locales highlighted several personal characteristics (e.g. female gender and low socioeconomic status) relating to higher levels of depression and anxiety following civil protest involvement. Moreover, the nature of involvement in the protests and the level of exposure to violence during the protests comprise important risk factors for mental health outcomes (Lau et al., 2016; Moussa et al., 2015; Ni et al., 2017; Norris et al., 2002; North & Pfefferbaum, 2013). However, much more information is needed to help identify those at a relatively higher risk for deleterious mental effects following participating in long civil protests.

One factor that may be related to such an effect is moral injury (MI; Litz et al., 2009). MI is an emerging construct that phenomenologically represents a particular trauma-related syndrome having biological, psychological, spiritual, behavioural, and interpersonal components. Jonathan Shay (1994) was the first to describe the idea of MI as ‘character wounds’ stemming from a leader's betrayal, using examples from Homeric literature and his clinical experience with US Army Vietnam veterans (Shay, 1994). Whereas construct validation efforts are being conducted, moral injury is generally assumed to result from exposure to events that involve either perpetrating or witnessing actions that violate one's core beliefs (Griffin et al., 2019; Litz et al., 2009) or betrayal by a leader or trusted authority (Shay, 1994). Exposure to one or more potentially morally injurious events (PMIEs) is perceived as a violation of deep moral beliefs by oneself or trusted individuals (Hall et al., 2022; Jinkerson, 2016).

Specifically, PMIEs were initially explored within a military context involving perpetration-based acts of commission (e.g. killing), omission (e.g. failing to prevent atrocities), bearing witness to perceived immoral acts, and experiencing the betrayal of trusted others (Litz et al., 2009; Litz & Maguen, 2012). However, non-military encounters, such as civil protests, riots, and revolutions, may expose civilians to severe moral and ethical challenges. While individuals can handle most of these events effectively, some PMIEs may lead to negative psychological outcomes among the protesters. For instance, direct perpetration of actions against police, failing to prevent violence, and witnessing immoral acts may transgress deeply held moral beliefs. In addition, some actions by a figure carrying legitimate authority, such as political leaders, may engender a sense of betrayal of what is right.

Nash et al. (2013) introduced the Moral Injury Events Scale (MIES) to study the consequences of PMIE exposure. Factor analyses have identified three subscales: PMIE-Self, PMIE-Others, and PMIE-Betrayal (Bryan et al., 2016). While the incidence of PMIEs varies between the era, the type of participants, and the nature of high-stake situations, initial MI research on military samples indicated that over a third of active-duty combatants and veterans report having been exposed to at least one type of PMIE (Griffin et al., 2019; Levi-Belz et al., 2023). However, in recent years, PMIEs have emerged as a topic of study in non-military populations, such as healthcare workers (Čartolovni et al., 2021), police personnel (Jones et al., 2020), teachers (Sugrue, 2020), and child protection professionals (Haight et al., 2017).

A single exploratory qualitative study was conducted on MI in civil protests (Karmel & Kuburic, 2021), finding that participants in Jordan’s ‘Arab Spring’ demonstrations reported exposure to all the PMIE dimensions. The authors stressed that PMIE-Betrayal (involving betrayal of what is right committed by a figure having legitimate authority, such as a political leader) in a high-stakes situation was widely experienced by Jordanians in the ‘Arab Spring’ protests (Karmel & Kuburic, 2021). However, to date, no quantitative study has empirically examined the prevalence of PMIE dimensions among protesters in civil demonstrations.

Several systematic reviews and meta-analyses have noted that PMIE exposure can heighten individuals’ vulnerability to psychiatric symptomatology (Griffin et al., 2019; Hall et al., 2022; McEwen et al., 2021; Williamson et al., 2018, 2021; Zerach et al., 2023). PMIEs have been shown to have particularly strong relationships with PTSS and depression (Levi-Belz et al., 2020).

Interestingly, both PTSS and depression were elevated mostly in the aftermath of exposure to the PMIE-Betrayal dimension (e.g. Levi-Belz et al., 2022; Zerach & Levi-Belz, 2018), which reflects an experience of betrayal of trusted commanders, authorities, and government leaders. These studies’ contributions notwithstanding, their findings are tempered by their focus on military or healthcare populations. No study to date has examined the PMIEs’ contribution to PTSS and depressive symptoms among protest participants.

1.1. The present study

Israel’s 2023 election gave rise to a far-right coalition that publicly declared its plans for a wide judicial overhaul aimed at changing the nature of Israel’s democratic regime (Albin et al., 2023). These changes include a judicial overhaul aimed at constraining the independence of Israel’s judiciary (e.g. Shultziner, 2023). Implementing these constitutional changes would effectively lead to a democratic backslide, as witnessed in recent years in countries such as Hungary, Poland, and Turkey. Following the coalition’s announcement, its plan has been met by unprecedented large-scale civil protests throughout the country, which have proceeded continuously for over 35 weeks in the form of demonstrations, riots, and various other non-violent resistance activities aimed at halting the coalition’s legislative efforts. The controversy surrounding the proposed legislation and the civil protest opposing it has led to a large-scale schism in Israel’s society (Albin et al., 2023; Shultziner, 2023).

In this study, conducted during August 2023 (between weeks 32 and 34 of the nationwide protests in Israel), we aimed to examine the mental effects of the judicial overhaul in Israel and the continued massive civil non-violent protest against it. Specifically, we aimed to understand the prevalence of PMIEs among civilians who participated actively in these protests. Furthermore, we sought to shed light on the specific contribution of the PMIE dimensions of self, other, and betrayal to PTSS and depression symptoms beyond the known risk factors for such deleterious psychological outcomes (Ni et al., 2020a). To our knowledge, this is the first study examining PMIEs and their psychological consequences among protesters. We posited two hypotheses:

1. Participants exposed to PMIEs during the protests will report higher levels of PTSS and depression than those not exposed to PMIEs.

2. PMIE exposure in general and PMIE-Betrayal exposure specifically will contribute to PTSS and depressive symptoms beyond demographics and protest-related characteristics (e.g. exposure to violence).

2. Methods

2.1. Participants

This cross-sectional design study sample comprised 4036 Israeli adult protesters who participated in the civilian demonstrations against the government’s proposed judicial reform in January–August 2023. Inclusion criteria were a minimum age of 18 and reporting that they were active in the protests from January to August in Israel. Among the consenting participants (n = 5274), 471 (8.9%) did not complete the study questionnaires, and 440 (8.4%) did not meet both inclusion criteria. Of the remaining 4363 participants, 325 (7.0%) completed only the sociodemographic, work, and protest-related questionnaires, whereas 4036 (92.5%) completed all or most of the questionnaires. A comparison of those who completed most of the questionnaire and those who completed only the demographic and protest-related characteristics parts did not reveal significant differences in most sociodemographic and protest-related variables. In summary, 4036 participants (75.9% females) completed all or most of the questionnaires; their demographic and protest-related characteristics are reported in Table 1.

Table 1. Characteristics of the study participants (N = 4036).

Variables		M (SD)	n (%)
Socio-demographic characteristics
Age		54.27 (12.45)
Education (years)		16.09 (1.81)
Gender	Male		24.0 (968%)
	Female		75.6 (3050%)
Country of birth	Israel		3614 (89.9%)
	FSU countries		80 (2.0%)
	America		156 (3.9%)
	Europe		111 (2.8%)
	Africa		27 (.7%)
	Other		32 (.8%)
SES	Far below average		204(7.6%)
	Below average		502 (15.0%)
	Average		708 (22.5%)
	Above average		1155 (36.8%)
	Far above average		537 (17.1%)
Marital status	Single		401 (10.0%)
	Married		2565 (63.6%)
	In relationship		553 (13.7%)
	Divorced		375 (9.3%)
	Widowed		95 (2.4%)
	Other		41 (1.0%)
Religiosity	Secular		3557 (88.5%)
	Traditional		381 (9.5%)
	Religious		70 (1.7%)
	Orthodox		13 (0.3%)
Protests-related characteristics
Involvement duration (months)		6.99 (1.60)
Extent of active involvement (protests per month)		5.41 (4.16)
Personal exposure to violence by the police	Not at all		3277 (83.7%)
	One time		386 (9.9%)
	Several times		209 (5.3%)
	Many times		43 (1.1%)
Being arrested by the police during protests	Not at all		3838 (98.3%)
	Once		49 (1.3%)
	Several times		18 (0.5%)
	Many times		5 (0.1%)
Involvement in active civil protests in the past	No		1946 (48.‏3%)
	Yes		1942 (48.2%)
	Not sure		145 (3.2%)

2.2. Measures

2.2.1. Moral injury events scale-civilian (MIES-C; Morriss & Berle, 2023)

The MIES-C is a 9-item self-report measure assessing possible events of moral injury among civilians. This scale is based on the Moral Injury Events Scale (MIES; Nash et al., 2013), designed primarily for investigating the military context. Compared with the original MIES, various instructions and items were modified in the MIES-C to remove military references. For example, ‘I feel betrayed by fellow service members who I once trusted’ was changed to ‘I feel betrayed by friends who I once trusted.’

The MIES-C taps exposure to perceived transgressions and comprises three subscales: (1) MIES-Self – four items assessing exposure to MI resulting from the respondent having committed acts or making decisions perceived by them to be morally wrong (e.g. ‘I acted in ways that violated my own moral code or values’); (2) MIES-Other – two items assessing exposure resulting from the respondent having witnessed or learned about others’ actions perceived by them to be morally wrong (e.g. ‘I am troubled by having witnessed others’ immoral acts’); and (3) MIES-Betrayal – three items assessing exposure to MI resulting from the respondent's perceiving being deceived or betrayed by others (e.g. ‘I feel betrayed by leaders whom I once trusted’). The nine statements were presented on a 6-point Likert-type scale, ranging from 1 (strongly disagree) to 6 (strongly agree). Thomas et al. (2023) demonstrated strong construct validity and positive associations with related constructs. For the current sample, the three subscales yielded high internal consistency: MIES-Self (α = .75), MIES-Other (α = .71), and MIES-Betrayal (α = .80).

2.2.2. International trauma questionnaire (ITQ; Cloitre et al., 2018)

The ITQ is a self-report measure for PTSD and complex PTSD. In this study, we focused on PTSD and, thus, used only the six items relevant to PTSD, reflecting three symptom clusters (re-experiencing, avoidance, and sense of threat). These six items (two items per symptom cluster) are presented on a 5-point Likert-type scale, ranging from 0 (not at all) to 4 (extremely). Moreover, three additional items measure functional impairment, which assess symptoms impact on (i) relationships and social life, (ii) work or ability to work, and (iii) other important aspects of life, such as parenting, school, or college work.

ITQ was operationalised in this study both as a dichotomised self-report ‘diagnosis’ of PTSD as a continuous variable and as a total symptom severity score of PTSS. To yield a PTSD diagnosis, one of each symptom cluster pair was required, as well as an indication of functional impairment in the last three items. Previous versions of the ITQ have been shown to be reliable and valid measures of PTSD (Shevlin et al., 2018), and the currently used version has recently been demonstrated to be a valid measure for PTSD diagnosis, based on the ICD-11 (Cloitre et al., 2018). For the continuous PTSS severity measure, a sum of symptoms was calculated, with the total score ranging between 0 and 24. For the current sample, the ITQ yielded high internal consistency for the PTSS (α = .86)

2.2.3. Patient health questionnaire-2 (PHQ-2; Kroenke et al., 2009)

The PHQ-2 two-item depression screening scale comprises the two DSM-5 diagnostic core criteria for depressive disorders (American Psychiatric Association, 2013): ‘little interest or pleasure in doing things’ and ‘feeling down, depressed, or hopeless.’ Equivalent to the full scale (PHQ-9), the PHQ-2 begins with the stem question: ‘Over the last 2 weeks, how often have you been bothered by the following problems?’ Items are presented on a 0–4 scale: ‘not at all’ (0), ‘several days’ (1), ‘more than half the days’(2), and ‘nearly every day’ (3). Thus, the total PHQ-2 sum scores range from 0 to 6. A score of ≥ 3 has been identified as a valid cut-off for differentiating between the normal range and probable cases of depression (e.g. Kroenke et al., 2003; Manea et al., 2016). The scale has been found to have high validity and a good resemblance to the full PHQ-9 scale. For the current sample, the PHQ-2 yielded good internal consistency (α = .78).

2.2.4. Exposure to violence during protests

To examine the extent of the participants’ exposure to violence during the protests, we formulated two measures tailored to this study: 1. Exposure to direct violence on oneself and 2. witnessing direct violence on other demonstrators. Exposure to direct violence was evaluated by three items focusing on violence inflicted by police and civilians (e.g. ‘To what extent were you a victim of physical violence by a police officer?’). Items are presented on a 4-point scale: 0 (not at all), 1 (one time), 2 (several times), and 3 (many times). A total score was calculated for each participant.

Witnessing violence on other protesters was evaluated by four items focusing on witnessing violence inflicted on other protesters by police and civilians (e.g. ‘To what extent were you a witness to physical violence toward other protesters by a police officer’). Items are presented on a 4-point scale: 0 (not at all), 1 (one time), 2 (several times), and 3 (many times). A total score was calculated for each participant.

2.2.5. Sociodemographic and protest-related characteristics

We collected various demographics, including age, gender, area of origin, socioeconomic status (SES), family status, religious orientation, and educational level. We also collected the participants’ protest-related characteristics, including duration of involvement in the civil protests (ranging from zero to eight months), amount of active involvement (number of attended demonstrations per month), and if they had previously been involved in active civil protests (Yes/No).

2.3. Procedure

Participants were recruited from online communities, online groups, and specialised social media forums and groups dedicated to the current civil protest during August 2023. The invitation post described the study’s aims as an ‘investigation of individuals’ experiences, emotional aspects, and attitudes regarding the civil protests.’ Those agreeing to participate received an explanation of the study's aims and a link to the related online survey through an online data-gathering website (in Qualtrics). Participants were required to affirm their willingness to participate and provide informed consent by their active participation. Ten vouchers (approximate value of each: US$50) were drawn among the participants who completed the questionnaires as compensation for their participation. Approval for this study was granted by the Ruppin Academic Center internal review boards.

2.4. Data analysis

In the first phase of data analysis, we calculated descriptive statistics, rates of probable self-reported PTSD and depression disorder diagnoses, as well as the prevalence of exposure to the PMIE dimensions. Following that, a multivariate analysis of variance (MANOVA) was performed to examine differences in PTSS and depressive symptoms between participants exposed to each PMIE dimension and those not (Yes/No).

In the second data analysis phase, the relationships between the study variables were examined using a series of Pearson correlation analyses. After calculating these intercorrelations, addressing the unique contribution of the PMIEs to PTSS and depressive symptoms, two four-step hierarchical regression analyses were conducted. All independent variables were centred before entering the regression models. To statistically control for the role of demographic characteristics of age, gender, and SES, these were inserted into the equation in Step 1. In Step 2, variables related to the level of involvement in the civil protests were entered. Exposure to violence and witnessing violence during the protests were entered in Step 3. Finally, after controlling for all other variables, the three PMIE dimensions were entered in Step 4 to examine the contribution of direct violence exposure and witnessing violence. The statistical significance of the coefficients was confirmed by constructing 95% confidence intervals using a bootstrap procedure (see Efron & Tibshirani, 1986) with 1000 resamples. The level of statistical significance was set at p = .05. SPSS (v26.0 for Windows) was used for all analyses.

3. Results

3.1. Prevalence of PTSD, depressive disorder, and PMIE exposure

Participants meeting the criteria of probable current PTSD and depression were determined by the validated cut-offs of the self-report diagnosis. The ITQ diagnostic criteria indicated that 10.6% of the participants (n = 426) suffered from probable current PTSD. The self-report diagnosis of depression, defined by the PHQ-2 cut-off point, was set as ≥ 3 (Manea et al., 2016). Accordingly, the prevalence of current depression was 44.6.% (n = 1786) of the participants, with 79.9% of them (n = 1427) reporting high or extreme daily functional difficulties due to these problems.

Regarding PMIE exposure, 96.6% (n = 3905) of the participants endorsed at least one MIES item tapping exposure to PMIEs at the agree or defiantly agree level. The most commonly endorsed MIES items were ‘I saw things that were morally wrong’ (91.7%), ‘I am troubled by having witnessed others’ immoral acts’ (70.5%), ‘I feel betrayed by leaders who I once trusted’ (59.5%), and ‘I feel betrayal by others outside my immediate circle who I once trusted’ (19.3%). Relating to the three PMIE dimensions, 15.0% (n = 607) of the participants endorsed at least one MIES-Self item at the agree or defiantly agree level, 92.7% (n = 3746) of the participants endorsed at least one MIES-Other item at the agree or defiantly agree level, and 63.9% (n = 2577) of the participants endorsed at least one MIES-betrayal item at the agree or defiantly agree level.

3.2. Differences between PTSS and depressive symptoms by PMIE endorsement

In this part of the analysis, we examined the differences in PTSS and depressive symptoms according to participants’ exposure to PMIEs of self, other, and betrayal. Based on responses to the MIES-C questionnaire (Morriss & Berle, 2023), we divided the participants into two groups reflecting the intensity of their PMIE exposure: Group 1 comprised those who endorsed at least one item of the relevant PMIE dimension at the agree or defiantly agree level, and Group 2 comprised those reporting not having been exposed to any of the three PMIE dimensions at the agree or defiantly agree level.

A MANCOVA analysis was carried out with the covariates being active protest involvement duration (in months) and amount of active involvement (demonstrations attended per month). The analysis examined differences in PTSS and depressive symptoms between the PMIE-exposed (Group 1) and the non-exposed (Group 2) participants. For PMIE-Self, the analysis revealed a significant group effect of measures, Wilks’ F approximation (2,3551) = 45.89, p < .001, Eta² = .04. The univariate ANCOVAs yielded significant group effects of PMIE-Self on PTSS F(1,3552) = 60.98, p < .001, Eta² = .03 and on depressive symptoms (1,3552) = 67.89, p < .001, Eta² = .03: those exposed to PMIE-Self (n = 607) were higher on both PTSS and depressive symptoms than those not exposed to PMIE-Self (n = 3429) (see Figure 1).

Figure 1. Posttraumatic stress symptoms as a function of exposure to potentially morally injurious events (PMIEs) among protesters (N = 4352).

For PMIE-Other, the analysis revealed a significant group effect of measures, Wilks’ F approximation (2, 3551) = 16.15, p < .001, Eta² = .04. The univariate ANCOVAs yielded significant group effects of PMIE-Self on both PTSS F(1,3552) = 29.83, p < .001, Eta² = .01 and depressive symptoms F(1,3552) = 13.47, p < .001, Eta² = .01. Those exposed to PMIE-Other (n = 3746) were higher on both PTSS and depressive symptoms than those who were not (n = 290). Figure 1 represents the differences regarding PTSS. For PMIE-Betrayal, the analysis revealed a significant group effect of measures, Wilks’ F approximation (2, 3551) = 86.51, p < .001, Eta² = .06. The univariate ANCOVAs yielded significant group effects of PMIE-Self on both PTSS F(1, 3552) = 143.37, p < .001, Eta² = .04 and depressive symptoms (1, 3552) = 94.07, p < .001, Eta² = .04, as those who exposed to PMIE-Self (n = 2577) were higher on both PTSS and depressive symptoms in compared to those who did not (n = 1459). Figure 1 represents the differences regarding PTSS.

3.3. The contribution of PMIEs to PTSS and depressive symptoms among protesters

3.3.1. Preliminary analysis

To test the hypothesis that PMIEs will contribute to demonstrators’ PTSS and depression, we first examined the intercorrelations between the study variables. As seen in Table 2, all three PMIE dimensions were significantly and positively correlated with both PTSS and depressive symptoms, with PMIE-Betrayal the dimension yielding the highest correlations. Moreover, all the PMIE dimensions were significantly and positively correlated with the two violence-related indicators: exposure to direct violence on self and witnessing violence on others during the demonstrations. As expected, the analysis yielded positive significant intercorrelations between exposure to violence, PTSS, and depressive symptoms.

Table 2. Intercorrelations between the study variables (n = 4352).

	1	2	3	4	5	6	7	8	9	10	11	12
1. Gender (Female = 1)	–
2. Age	.00	–
3. SES (1–5)	−.22***	.13**	–
4. Involvement duration (# months)	−.01	.17***	.10***	–
5. Extent of active involvement (# demonstrations per month)	.04**-	.04**	.05**	.33**	–
6. Exposure to violence in protests	−.08***	−.09***	−.01	.20***	.28***	–
7. Witnessing violence in protests	.06***-	−.14***	−.00	.26***	.38***	.58***	–
8. PMIE-Self	.04*	−.13***	−.07**	.06**-	.06**-	.04*	.03***	–
9. PMIE-Other	.08***	.08***	−.01	.08***	.05***	.12***	.14***	.09***	–
10. PMIE-Betrayal	.05***	.01	−.03	.02	.05**	.19***	.15***	.16***	.18**	–
11. Depression	.15*	−.04*	−.06**	.05**	.07***	.14***	.14***	.13***	.12***	.24***	–
12. PTSS	.06*	.03	−.05**	.09***	.14***	.26**	.25***	.13***	.17***	.31***	.41***	–
M	–	54.30	3.‏24	6.99	5.41	4.44	7.01	7.25	10.46	9.39	2.56	6.42
SD	–	12.09	1.15	1.60	4.16	1.70	2.28	3.80	2.45	3.67	1.71	4.83
Range	–	18–96	1–5	1–8	1–30	0–10	0–10	4–24	2–12	3–18	0–6	0–24

Note: *p < .05; **p < .01; ***p < .001. Gender: Male = 0, Female = 1; SES = Socio-economic status; PMIE = potentially morally injurious event.

Table 2 shows that several demographic variables of gender and SES were related significantly to both outcome measures, as female gender and lower SES were associated with higher levels of both PTSS and depressive symptoms. Furthermore, we found significant and positive correlations between involvement duration and amount of time spent in demonstrations on the one hand and PTSS and depression levels on the other.

3.3.2. Hierarchical regression analyses

To fully understand the unique contribution of PMIE exposure to PTSS and depressive symptoms beyond the contribution of the other study variables, we conducted two four-step hierarchical regression analyses (Cohen et al., 2003), one with PTSS as an outcome measure and the other with depressive symptoms as an outcome measure. The same independent variables appeared in the two analyses. Gender, age, and SES were used as covariates in both models and entered in Step 1. In Step 2 and Step 4, protest-related characteristics of involvement and exposure to violence were entered, respectively. In the final step, the three PMIE dimensions of self, other, and betrayal were entered to examine their contribution to PTSS and depressive symptoms after all other variables were entered.

Overall, the model for PTSS as an outcome measure was found significant, accounting for 18.2% of the variance of current PTSS, F_model (10,2894) = 61.31, p < .001. Specifically, as seen in Table 2, in Step 1, demographic variables significantly contributed 1% to PTSS levels, F_change-Step1 (3, 2901) = 6.96, p < .001. The protest involvement variables, entered in Step 2, significantly contributed to PTSS levels and added 2% to the explained variance (R-squared change) in PTSS, F_change-Step2 (2, 2901) = 31.92, p < .001. Violence exposure variables, entered in Step 3, also significantly contribute to PTSS levels, adding 7% to the explained variance (R-squared change) in PTSS, F_change-Step3 (2, 2901) = 108.28, p < .001. Finally, the PMIE dimensions, entered at Step 4, predicted PTSS levels, F_change-Step4 (3, 2901) = 87.51, p < .001, adding 8% to the explained variance of PTSS, beyond all other predictors. While all PMIE dimensions contributed significantly to PTSS, it should be noted that the strongest predictor was PMIE-Betrayal, β = .31, p < .001.

The model for depressive symptoms as an outcome measure was also found significant, accounting for 11.3% of the variance of current depressive symptoms, F_model (10,2894) = 34.56, p < .001. As seen in Table 2, in Step 1, demographic variables significantly contributed 2% to depressive symptoms, F_change-Step1 (3,2901) = 23.43, p < .001. Involvement variables, entered in Step 2, significantly contributed to PTSS levels, adding 3% to the explained variance (R-squared change) in depressive symptoms, F_change-Step2 (2, 2089) = 12.68, p < .001. Violence exposure variables, entered in Step 3, also significantly contribute to PTSS levels, adding 3% to the explained variance (R-squared change) in depressive symptoms, F_change-Step3 (2,2087) = 34.45, p < .001. Finally, PMIE dimensions, entered at Step 4, predicted depressive symptoms levels, F_change-Step4 (3,2896) = 60.07, p < .001, adding 6% to the explained variance of depressive symptoms beyond all other predictors. Whereas all PMIE dimensions contributed significantly to depressive symptoms, it is noteworthy that PMIE-Betrayal was the strongest predictor, β = .19, p < .001 (Table 3).

Table 3. Summary of Hierarchical Regression Coefficients of depressive symptoms among protestors by demographics, violence exposure and potentially moral injurious events (N = 4352).

	Outcome measures
Predictor variables	PTSS			Depressive symptoms
	B	t	β	B	t	β
Step 1
Gender	.58	2.66	.05**	.56	7.45	.14***
Age (years)	.01	2.28	.04*	.00	−1.87	−.03
SES	−.20	−2.57	−.04*	−.04	−1.45	−.03
R^2 (ΔR^2), Fchange	.01 (.01), Fchange = 6.96, p < .001			.02 (.02), Fchange = 23.43, p < .001
Step 2
Involvement duration (in months)	.16	2.65	.05**	.05	2.17	.04*
Amount of active involvement (demonstrations per month)	.14	6.27	.12***	.03	3.67	.07***
R^2 (ΔR^2), Fchange	.03 (.02), Fchange = 31.92, p < .001			.03 (.01), Fchange = 12.68, p < .001
Step 3
Exposure to violence in protests	.53	8.38	.19***	.10	4.38	.10***
Witnessed violence in protests	.24	6.07	.14***	.05	3.43	.14***
R^2 (ΔR^2), Fchange	.10 (.07), Fchange = 108.28, p < .001			.05 (.02), Fchange = 34.45, p < .001
Step 4
PMIE-Self	.11	4.92	.09***	.04	4.63	.08***
PMIE-Other	.17	4.33	.08***	.03	2.52	.05***
PMIE-Betrayal	.31	12.92	.23***	.09	10.37	.19***
R^2 (ΔR^2), Fchange	.18 (.08), Fchange = 87.51, p < .001			.11 (.06), Fchange = 60.07, p < .001

Note: *p < .05; **p < .01; ***p < .001; PMIE = potentially morally injurious event.

4. Discussion

Involvement in civil protest has the potential to significantly affect protesters demonstrating against government actions. Thus, this study’s primary aim was to shed light on the mental burden (i.e. PTSS and depression) endured by protesters during the non-violent protest against the Israeli government’s judicial overhaul. Moreover, we aimed to understand the role of exposure to protest related PMIEs in exacerbating the demonstrators’ mental burden in these unprecedented times. To our knowledge, this is the first study to examine PMIEs and their consequences in protest situations.

The results of this study highlight that demonstrators in our sample experienced a high mental burden during the eight months of the ongoing protests. Of the sample, 10.6% were found to have a probable self-report PTSD diagnosis and close to half (44.6%) were found to have a probable self-report diagnosis of depression. The prevalence of PTSD aligns with several studies of demonstrators internationally (e.g. Ni et al., 2020a; Ojeahere et al., 2021); however, the prevalence of depression among the Israeli protesters appears to be higher than for most international samples (44.6% in the Israeli sample compared with an international average of 10%, Ni et al., 2020a), save Hong Kong demonstrators, who showed similar depression rates (Ni et al., 2017). Considering that 75.9% of the Israeli sample reported major daily functional difficulties due to depressive symptoms, our findings indicate that the perceived threat to democracy and civil protests against it put Israeli protesters at risk for psychopathology. This high prevalence of dysfunctionality may be due to the extensive duration of Israel’s current protests (approximately 240 days compared with 8–79 days in most of the reported studies; Ni et al., 2020a). This prevalence underscores the importance of understanding the specific risk factors associated with heightened psychological difficulties experienced by civil protesters in Israel.

Importantly, we found that transgressive acts, particularly PMIE-betrayal, were highly correlated to the protesters’ PTSS and depressive symptoms. Whereas these results should be considered cautiously due to the study’s cross-sectional design, they suggest that PMIEs may comprise a unique risk factor for PTSS and depression among protesters participating in the demonstrations against the Israeli government’s proposed judicial overhaul. The notion that exposure to profound ethical and moral dilemmas is related to psychopathology and psychological difficulties has been posited conceptually (e.g. Litz et al., 2009) and substantiated by empirical studies on military personnel (Schwartz et al., 2022; Zerach & Levi-Belz, 2018) and healthcare workers (Čartolovni et al., 2021; Zerach & Levi-Belz, 2022). The current empirical study, however, extends a recent qualitative study (Karmel & Kuburic, 2021) and is the first to quantitatively demonstrate the contribution of PMIE exposure to the protesters’ mental burden beyond known risk factors such as age, gender, and exposure to violence. Thus, these findings emphasised that moral dilemmas, primarily the experience of betrayal, are highly relevant to the protest-related psychological distress among protesters.

The betrayal experience comprises a high-magnitude moral injury as it relates to deleterious actions performed by trusted leaders or authority figures (Frankfurt & Frazier, 2016). In the context of civil movements and demonstrations, the betrayal experience may be more related to political authority figures such as government ministers, who were perceived as betraying civilian moral and ethical expectations about taking responsibility for the event(s), acting in moral ways, maintaining a commitment to the nation’s core values, and promoting a balanced policy that addresses the broad needs of all citizen groups (Karmel & Kuburic, 2021; Shay, 1994). Some studies have suggested that betrayal experiences may place individuals at risk for perpetrating other transgressive acts and increasing their vulnerability to adverse consequences (Griffin et al., 2019; Levi-Belz et al., 2020). Upon considering military personnel, Shay (1994) speculated that betrayal perpetrated by commanding authorities corrodes the cohesion and effectiveness of military units and the safety and security of combat personnel. In studies among healthcare workers during COVID-19 (e.g. Benatov et al., 2022; Hegarty et al., 2022; Weber et al., 2023), betrayal experiences were found to shatter the inner structure within the hospital wards, increase burnout, and double the sense of uncertainty among healthcare workers, which had already peaked due to the COVID-19 circumstances (Levi-Belz & Zerach, 2022).

We can suggest that among those protesting against the judicial overhaul in Israel, the country’s coalition-led efforts are experienced as a betrayal of the nation’s core values and consensual standards by the very leaders entrusted with protecting these values and standards. This experience of betrayal is what can lead to a breakdown in citizens’ basic trust in their government and elected leadership (Albin et al., 2023). Such a breakdown in basic trust may lead to detrimental effects on mental health. As leaders have been described as attachment figures whose functioning can have diverse effects on their followers (Davidovitz et al., 2007), political leaders’ moral betrayal could be experienced psychologically as a betrayal by a parental figure and, thus, engendering a harmful emotional experience. Betrayal by attachment figures has been linked repeatedly to trauma (e.g. Hocking et al., 2016; Owen et al., 2012) and thus can explain the much higher rates of PMIE-Betrayal and PMIE-Others found in our study.

The current findings should be considered in light of several limitations. First, as a cross-sectional designed study, causality between variables cannot be assumed. For instance, some of the study’s participants who reported high PTSS or depressive symptoms may have suffered from psychological difficulties prior to their involvement in the protests. Moreover, it is even plausible to suggest that those who suffer from depression symptoms may have been relatively more attracted to involvement in the protest or encouraged by others to be involved in a massive social experience. However, our sample’s high rate of psychological difficulties and the significant relationships between the level of protest involvement and psychological distress suggest that, to some extent, these difficulties are linked to their participation in the protests. Future studies should also examine the longitudinal contribution of PMIEs to PTSS and depression among protesters.

Second, we employed a non-representative volunteer sample of protesters. The sizable sample may help reduce this limitation, but cultural and societal characteristics should be considered before generalising these findings to protesters from different nations and situations. For example, as the current protests in Israel are of much longer duration than most other reported protests (e.g. Ni et al., 2020a), their duration may partially explain the relatively higher depressive symptoms and the higher PMIE prevalence found in our study. A further constraint to the findings’ generalizability relates to the sample’s underrepresentation of some sub-groups, such as males, which may impair the study's external validity. Lastly, PTSS, depression, and PMIEs were assessed using participants’ subjective accounts rather than by clinical interviews of transgressive acts. Thus, the measures may be biased due to social desirability factors. However, PMIEs result from a perceived clash with individuals’ inner moral code rather than their behaviour, thus lending legitimacy to subjective self-reports of PMIEs.

4.1. Conclusions and implications

During unprecedented civil protests against the Israeli government’s judicial overhaul, protesters have experienced complex challenges. These challenges include first of all the actions of the government themselves, which may be a strong factor in the development of mental health difficulties. Other challenges may be also the hopelessness experience accompanied to the long period of protests which did not change the government actions, and the exposure to violence and involvement in mass demonstrations over a lengthy period, which may also contribute to the psychological burden. Thus, it is not surprising that protesters suffered high rates of psychopathology, such as PTSD and depression. Importantly, among the risk factors for these psychopathologies are protest-related PMIEs, found to be important contributors to PTSS and depressive symptoms beyond other known demographic and protest-related characteristics.

Several implications can be highlighted in light of these results. First, during the period of the protest, protesters should be encouraged to seek informal support such of family members and feirends (e.g. Ni et al., 2020b; Williamson et al., 2021), which has been shown to be an effective buffer between stress and psychopathology in the aftermath of PMIE experiences (Levi-Belz et al., 2020). Moreover, clinicians should also be aware that protesters are subject to a high risk for psychopathology during the protest, both generally and specifically as a result of PMIE experiences. As individuals who develop moral injury-related mental health disorders are often reticent to speak about guilt, shame, and disgust (Griffin et al., 2019; Litz et al., 2009; Williamson et al., 2021), clinicians should make sufficient sensitive inquiry about PMIEs among protesters presenting mental health difficulties during the civil protest.

Second, the high percentages of betrayal experiences found among protesters in our sample and their significant contribution to psychological difficulties underscore the importance of active screening for PMIE experiences among them. If evidence of PMIE experiences is presented, especially PMIE-Betrayal, psychological support for the affected protesters should be prioritised and made more readily accessible.

Third, the findings highlight the critical role of political leaders, managers, employers, and governments in supporting protesters’ physical and emotional health in their daily activities during the unfolding protest situations. Examples of such actions might include a public acknowledgment and recognition of the protesters’ stress and pressure (Roycoft et al., 2020). In light of the high proportion of protesters reporting exposure to PMIE-Other and PMIE-Betrayal, it can be suggested that protest movement leaders speak about the betrayal experience experienced by the civilians and thus, help normalise these experiences, which, in turn, could help protesters obtain a deeper grasp of the situation they are confronting and even seek counselling when the distress disturbs their functioning. Other ways to build resilience at the system level may include developing strategies to improve the ethical climate to facilitate the community protest leaders’ ability to support the protesters in dealing with ethical issues (Maffoni et al., 2020). Furthermore, protest leaders can actively promote a belongingness experience among the protesters, as belongingness has been found to buffer the influence of PMIEs on distress (Benatov et al., 2022).

Data availability

The data that support the findings of this study are available from the corresponding author, [YLB], upon reasonable request.

Disclosure statement

No potential conflict of interest was reported by the author(s).