Moral injury and pre-deployment personality factors as contributors to psychiatric symptomatology among combatants: a two-year prospective study

ABSTRACT

Background: Combatants who are exposed to events that transgress deeply held moral beliefs might face lasting psychopathological outcomes, referred to as Moral Injury (MI). However, knowledge about pre-deployment factors that might moderate the negative consequences of MI is sparse. In this prospective study, we examined pre-enlistment characteristics and pre-deployment personality factors as possible moderators in the link between exposure to potentially morally injurious events (PMIEs) and psychiatric symptomatology among Israeli active-duty combatants.

Methods: A sample of 335 active-duty Israeli combatants participated in a 2.5-year prospective study with three waves of measurements (T1: 12 months before enlistment, T2: Six months following enlistment – pre-deployment, and T3: 18 months following enlistment – post-deployment). Participants’ characteristics were assessed via semi-structured interviews (T1) and validated self-report measures of personality factors: emotional regulation, impulsivity, and aggression (T2) and combat exposure, PMIEs, psychiatric symptomology and posttraumatic symptoms (T3) between 2019 and 2021.

Results: Pre-enlistment psychiatric difficulties and negative life events contributed to higher exposure to PMIEs post-deployment. Higher levels of pre-deployment aggression and lower levels of emotional regulation and impulsivity moderated the association between betrayal, PMIEs and psychiatric symptomology post-deployment, above and beyond pre-enlistment psychiatric difficulties and life events.

Conclusions: Our results highlight that pre-deployment emotional regulation, impulsivity and aggressiveness levels should be assessed, screened, and identified among combatants, as they all facilitate psychiatric symptomology (and PTSS) after combatants are exposed to PMIEs of betrayal. Such pre-assessment will enable the identification of at-risk combatants and might provide them with tailor-made preparation regarding moral and ethical situations that should be investigated in future research.

HIGHLIGHTS

• Pre-enlistment psychiatric difficulties contributed to higher exposure to PMIEs post-deployment.

• Pre-deployment personality moderated the associations between PMIEs and psychiatric symptomatology among combatants.

• Higher levels of pre-deployment emotional regulation moderated the association between betrayal PMIEs and post-deployment PTSS symptoms.

Antecedentes: Los combatientes que están expuestos a eventos que transgreden creencias morales profundamente arraigadas pueden enfrentar resultados psicopatológicos duraderos, efectos conocidos bajo el nombre de daño moral (MI, por sus siglas en inglés). Sin embargo, el conocimiento sobre los factores previos al despliegue que podrían moderar las consecuencias negativas del MI es escaso. En este estudio prospectivo, examinamos las características previas al alistamiento y los factores de personalidad previos al despliegue como posibles moderadores en la relación entre la exposición a eventos potencialmente dañinos moralmente (PMIEs, por sus siglas en inglés) y la sintomatología psiquiátrica entre los combatientes israelíes en servicio activo.

Métodos: Una muestra de 335 combatientes activos participaron en un estudio prospectivo de 2.5 años con tres oleadas de mediciones (T1: 12 meses antes del alistamiento, T2: 6 meses después del alistamiento-pre-despliegue, y T3: 18 meses después del alistamiento-pos-despliegue). Las características de los participantes se evaluaron mediante entrevistas semiestructuradas (T1) y medidas de auto-reporte validadas de factores de personalidad: incluyendo regulación emocional, impulsividad y agresión (T2) y exposición al combate, PMIEs, sintomatología psiquiátrica y síntomas postraumáticos (T3) entre los años 2019-2021.

Resultados: Las dificultades psiquiátricas previas al alistamiento y los eventos de vida negativos contribuyeron a una mayor exposición a los PMIEs después del despliegue. Los niveles más altos de agresión antes del despliegue y los niveles más bajos de regulación emocional e impulsividad moderaron la asociación entre la traición, los PMIEs y la sintomatología psiquiátrica posterior al despliegue, más allá de las dificultades psiquiátricas previas al alistamiento y los eventos de vida.

Conclusiones: Nuestros resultados resaltan que los niveles de regulación emocional, impulsividad y agresividad previos al despliegue deben evaluarse, examinarse e identificarse entre los combatientes, ya que todos ellos facilitan la sintomatología psiquiátrica (y PTSS) después de que los combatientes están expuestos a los PMIEs de traición. Dicha evaluación previa permitirá identificar a los combatientes en riesgo y podría brindarles una preparación personalizada en relación a situaciones de materia moral y ética que deberían ser investigadas en futuras investigaciones.

KEYWORDS:

• Moral injury
• pMiEs
• personality
• aggressiveness
• prospective

PALABRAS CLAVE:

• Daño moral
• PMIEs
• personalidad
• agresividad
• prospectivo

1. Introduction

Engaging in modern warfare and guerrilla combat within both combatant and civilian settings might expose combatants to severe moral and ethical challenges. Recent studies have defined Moral Injury (MI) as the exposure to transgressive harms and the negative outcomes of those experiences (Litz & Kerig, 2019). Thus, besides the possible exposure to potentially traumatic events (PTEs) during combat, combatants might be exposed to one of two broad dimensions of Potentially Morally Injurious Events (PMIEs): PMIEs committed or omitted by oneself and being exposed directly or indirectly to others’ transgressive acts (such as betrayal by significant others; Jordan et al., 2017). Although the prevalence of PMIEs is likely to vary by branch, theater, era and military status, between 10% and 30% report exposure to PMIEs during military service (e.g. Wisco et al., 2017).

Theoretical models (e.g. Farnsworth et al., 2017; Litz et al., 2009) have postulated that exposure to PMIEs that transgress core values in high stake situations may cause significant moral dissonance. Among a significant minority of combatants, an unresolved moral dissonance may lead to severe moral pain, MI-related experiences of guilt and shame, as well as mental and behavioural health outcomes. Recently, several systematic reviews and meta-analyses have shown that exposure to PMIEs during military service can lead to increased vulnerability to psychiatric symptomatology, such as posttraumatic stress symptoms (PTSS), depression, anxiety, substance abuse and even suicidal ideation and behaviour (Hall et al., 2022; McEwen et al., 2021). However, given the heterogeneity of combatants’ subjective experience of transgressive acts, it is important to examine the unique contribution of personal characteristics and personality factors to combatants’ exposure to PMIEs, PTSS and psychiatric difficulties.

Over the last decade, only a handful of studies have examined which combatants’ risk factors are associated with exposure to PMIEs in military settings. For example, one study found associations between pre-enlistment personal characteristics such as younger age, having no more than a high school education, and exposure to negative life events such as physical abuse during childhood or having had a family member arrested, with harm to civilians or prisoners in the Vietnam War (Dohrenwend et al., 2013). Following Litz et al.’s (2009) MI working model, some studies highlight the contribution of pre-deployment personality factors such as anger and aggressiveness (Maguen et al., 2012) and lack of constraint (tendency toward impulsivity; Holowka et al., 2012) to the commitment of transgressive acts. The most consistent finding is the association between combat exposure and the likelihood of exposure to PMIEs (Frankfurt & Frazier, 2016). The generalisation of these findings is somewhat questionable due to their cross-sectional and retrospective designs, which limit the ability to examine temporal sequencing between antecedents and outcomes.

The associations between risk factors, PTSS and psychiatric symptomatology following military service are far more firmly established. For example, a recent umbrella review of systematic reviews and meta-analyses found that socio-demographic factors (e.g. female gender, pre-trauma factors (e.g. family history of psychiatric disorder), peri-trauma factors (e.g. trauma severity), were associated with PTSD (Tortella-Feliu et al., 2019). Unfortunately, however, most studies rely exclusively on cross-sectional or retrospective designs. A review of prospective studies that examined the contribution of pre-trauma risk factors in the etiology of PTSD found six categories: (1) cognitive abilities (e.g. low I.Q.); (2) coping and response styles (e.g. avoidance); (3) personality factors (e.g. high negative affect); (4) psychopathology (e.g. PTSS); (5) psychophysiological factors (e.g. high arousal); and (6) social-ecological factors (e.g. social support) (DiGangi et al., 2013).

To provide an overarching structure for selecting pre-trauma personality factors, we adopted the three-factor model of personality. This model is comprised of three higher-order dimensions: Positive emotionality/Extraversion (PEM), which represents positive emotions and tendencies toward active involvement; Negative emotionality/Neuroticism (NEM), which represents negative emotion and a tendency toward adversarial interactions with others; and Constraint/Inhibition (CON) which represents restraint vs. recklessness. These factors are considered relatively stable, heritable and have been extensively investigated regarding PTSS and psychiatric symptomatology (Jakšić et al., 2012; Miller, 2003). In the present study, we focused on aggressiveness as a facet of NEM and affect dysregulation and impulsivity as facets of CON. Indeed, it has been found that affect dysregulation and high impulsivity (Pencea et al., 2020) and aggressiveness (high hostility and anger; van Zuiden et al., 2011) are associated with PTSD. The present study will prospectively examine the direct contribution of pre-enlistment personal characteristics and pre-deployment personality risk factors to post-deployment PTSS and psychiatric symptomatology following exposure to PMIEs during military service assignments.

Importantly, there is a dearth of knowledge regarding risk factors that could explain the conditions under which PMIEs may contribute to psychopathological sequelae (Hall et al., 2022). For example, a few cross-sectional studies found that tendency to overidentify with one’s failings and shortcomings (Kelley et al., 2019), higher levels of rumination (Hamrick et al., 2020), and low/average intolerance of uncertainty (Zerach & Levi-Belz, 2019) strengthen associations between PMIEs and suicidality. Emotional dysregulation was found to moderate the association between exposure to PMIEs and alcohol misuse among veterans (Forkus et al., 2021). Other prospective studies among veterans found that pre-deployment hostility (van Zuiden et al., 2011) or negative affectivity (Rademaker et al., 2011) strengthen associations between general combat exposure and PTSD. To date, no prospective studies have examined pre-deployment personality risk factors as moderators of the links between exposure to PMIEs and psychiatric symptomatology. Following Litz et al.’s (2009) assumption regarding negative affectivity, which may hinder self-forgiveness and increase self-censorship and moral injury, and the moderating role of CON variables to PTSD (Miller, 2003), we suggest that pre-deployment high aggressiveness and low ability for emotional regulation would moderate the links between exposure to PMIEs and psychiatric symptomatology.

1.1. The present study

To sum up, there are no prospective data regarding the antecedents and outcomes of exposure to PMIES and its psychiatric sequelae among active-duty combatants in general and non-U.S. combatants in particular. To fill these gaps, our goal was to examine prospectively the links between pre-enlistment characteristics and pre-deployment personality factors with exposure to PMIEs and psychiatric consequences among Israel Defense Forces (IDF) active-duty combatants.

IDF combatants’ routine missions include peacekeeping tasks as well as operations in urban and civilian environments in the West Bank (alongside the traditional missions of securing Israel’s borders and participating in traditional warfare). These military assignments render them a highly relevant population, as their missions may expose them to ethical challenges, especially if non-combatant civilians are involved (Gelkopf et al., 2016).

Operationally, in this study, we conducted multi-stage prospective research in which we followed active-duty combatants over their first 18 months of service at three measurement points: T1 – pre-enlistment information from computerised records at the IDF Military Induction Center (MIC), including participants’ cognitive index score and performance prediction; T2 – the pre-deployment point, six months following enlistment at basecamps (negative life events, psychiatric symptomatology and PTSS pre-enlistment), and pre-deployment personality factors (emotional regulation, impulsiveness, aggression); and T3 – post-deployment (18 months following enlistment) where we gathered information on exposure to PMIEs and combat induced psychopathology (PTSS and psychiatric symptomatology) in the aftermath of initial deployment.

In this exploratory study, we sought to answer the following questions: (1) what is the contribution of pre-enlistment characteristics and pre-deployment personality factors to exposure to PMIEs; (2) what are the contributions of pre-enlistment characteristics, pre-deployment personality factors, and exposure to PMIEs to PTSS and psychiatric symptomatology, and (3) what are the moderating roles of pre-deployment personality factors in the associations between exposure to PMIEs, to PTSS and psychiatric symptomatology post-deployment.

2. Method

2.1. Participants

The current sample included 335 active-duty combatants from one of the IDF combat brigades. This particular brigade was chosen due to its intensive operational duties, both within Israeli borders and among civilians in the West Bank of Israel. Inclusion criteria required participants to have served as active combatants and to have completed all three measurement points. Of all participants who gave their consent in T2 (N = 547), 335 (61.2%) participants provided data at all three measurement points. Four hundred and forty-six participants completed only one measurement: 215 participants with data on T2 without T3 and 231 participants with data on T3 without T2; all participants had data on T1. Significant differences in marital status were found between those who completed all measurements and those who did not (χ²(2) = 10.55, p = .00). A higher number of participants who completed all measurements (4.2%) were married, as compared with none among participants with only one measurement. No significant differences were found between the groups on other socio-demographic variables, pre-enlistment variables (T1) and main outcomes variables (T2).

We note that all participants in the study were in the same age cohort and on mandatory service in the IDF. All participants were males; their mean age was 19.42 (SD = 1.11), meaning that most participants were 19 at T2. On average, they had completed 12.01 (SD = .45) years of education. Most participants were Israeli-born (n = 278, 93.9%) and single (n = 279, 94.2%). In terms of religiosity, 43% (n = 138) were religious, 26.3% (n = 84) were traditional, and 23.1% (n = 74) were secular.

2.2. Measures

2.2.1. T1 – pre-enlistment (one year prior to military service)

We collected pre-enlistment personal characteristics through the Military Induction Center’s (MIC) computerised records. Specifically, data were collected regarding participants’ cognitive index score and performance prediction, derived from a standard semi-structured interview conducted by trained psychology technicians and designed to predict soldier performance (Gal, 1986; see also Levi-Belz et al., 2018).

Cognitive Index (CI): an intelligence evaluation score, a highly valid measure of general intelligence, equivalent to a normally distributed I.Q. The CI comprises four subtests: Arithmetic and Similarities, which are similar to the equivalent subtests from the Wechsler Intelligence scales; Raven’s Progressive Matrices (a measure of nonverbal abstract reasoning and problem-solving abilities); and the Otis Test of Mental Ability, which measures the capacity to understand and carry out verbal instructions. CI scores range from 10 to 90, with a mean of 50 (Levi-Belz et al., 2018).

Performance Prediction Score (PPS): a composite score derived from the interview score and the empirically weighted scores on several indexes such as the cognitive index, combat suitability, years at school, and command of the Hebrew language. The scale score ranges from 42 to 56, with higher scores indicating a higher quality of soldier and combat capability (Gal, 1986).

2.2.2. T2 – pre-deployment (6 months following enlistment)

In this wave, we assessed pre-deployment personality factors using three validated self-report measures: the Emotional Regulation Questionnaire (ERQ; Gross & John, 2003), the Barratt Impulsiveness Scale (BIS, Barratt, 1959) and the Buss-Perry Aggression Questionnaire (AGQ; Buss & Perry, 1992). Furthermore, for statistical control, we assessed negative life events before enlistment with the Life Events Checklist (LEC-5; Weathers et al., 2013), the Brief Symptoms Inventory (BSI; Derogatis, 1977) and the Posttraumatic Stress Disorder Checklist (PCL-5; Blevins et al., 2015) for events before enlistment.

2.2.3. T3-Post-deployment (18 months following enlistment)

In this wave, we assessed combat exposure, PMIEs and psychiatric symptomatology. Combat exposure was measured by the Combat Experiences Scale (CES; Hoge et al., 2004). Exposure to PMIEs was assessed by the Moral Injury Event Scale (MIES; Nash et al., 2013), tapping exposure to perceived transgressions committed, comprising three subscales: PMIE-Self, PMIE-Others, and PMIE-Betrayal. Importantly, whereas most studies have validated the use of the MIES to capture exposure to transgressions (e.g. Bryan et al., 2016; Nash et al., 2013), others have noted that this measure could also measure subjective distress following exposure to transgressions acts (Griffin et al., 2019).

Psychiatric symptomatology was measured by two validated self-report measures: Global severity index was measured by the Brief Symptoms Inventory (BSI; Derogatis, 1977), and PTSS was measured by the Posttraumatic Stress Disorder Checklist (PCL-5; Blevins et al., 2015).

2.3. Procedure

The recruitment process for the first face-to-face measurement (T2; six months after enlistment) comprised two stages: In the first phase, explanatory sessions were held for the combatants at the Kfir Brigade’s divisional training base without the presence of commanders. The explanations emphasised that the study includes two sessions for completing questionnaires – the first to be completed currently and the second after about a year. Additionally, participant information from the IDF recruitment bureau will be obtained. We emphasised that consent to participate would have no effect on their continued military service in any way. At the conclusion of these meetings, the combatants received a written explanation and an informed consent form relating to both parts of the study.

The meetings were scheduled on days before vacations so that combatants could consult with family and relatives regarding their participation. In the second stage, joint group meetings were scheduled for combatants who had agreed to participate in the study. These combatants were required to reaffirm their willingness to participate and to sign an informed consent form, and then they completed the T3 questionnaires. For participants who completed T2, computerised records from the Military Induction Center (MIC) were obtained (T1, collected from IDF candidates approximately one year before enlistment) was obtained. T2 participants were invited again to continue taking part in the study at the T3 face-to-face measurement wave (18 months following enlistment), held at the Kfir Brigade battalion bases. Those who agreed reaffirmed their informed consent to participate. Ethical approval for this study was granted by the IDF Medical Corps Institutional Review Board (Helsinki Committee).

2.4. Statistical analysis

First, we computed descriptive statistics of demographic and questionnaire data. Second, the associations between the study variables were examined with a series of Pearson correlation analyses. Third, to address the unique contribution of the independent variables at T1 and T2 to PMIE dimension levels, three four-step hierarchical regression analyses were conducted with PMIE-Self, PMIE-Other, and PMIE-Betrayal as dependent variables (T3). All independent variables were standardised before entering them into the regression models. We followed a chronological rationale for the order of entering variables into the regression models. In the first step of each regression, we entered measures of pre-enlistment negative life events and psychiatric symptoms for statistical control. In the second step, we entered the participants’ personal characteristics – their pre-enlistment cognitive index score and performance prediction – as derived from the MIC computerised records (T1). In the third step, we entered the personality factors of emotional regulation, impulsivity, and aggression, which were measured pre-deployment (T2), and in the final step, we entered combat exposure experiences (T3).

Fourth, to address the unique contribution of the independent variables at T1 and T2 to the psychiatric symptomatology of GSI and PTSS at T3, two 6-step hierarchical regression analyses were conducted: BSI-T3 and PTSS-T3 as dependent variables. All independent variables were standardised before being entered into the regression models. We followed a chronological rationale for the order of entering the variables into the regression models. In the first step of each regression, for statistical control, we entered measures of negative life events and the pre-enlistment GSI/PTSS scores. In the second step, we entered the pre-enlistment cognitive index and performance prediction scores obtained from the MIC computerised records (T1). In the third step, we entered the personality factors of emotional regulation, impulsivity, and aggression, which were measured pre-deployment (T2), and in the fourth step, we entered combat exposure experiences (T3). The fifth step contained the measures of PMIE-Self, PMIE-Other and PMIE-Betrayal (T3). In the final step, we entered all two-way interactions between the three personal factors (emotional regulation, aggression, and impulsivity) and the three PMIE dimensions.

Last, to examine the possible moderation of the mentioned links, ordinary least squares regression analysis was conducted using the PROCESS macro in SPSS (Hayes, 2013, Model 1). We tested the significance of interaction effects with a pick-a-point approach for probing moderation effects. This approach involves selecting representative moderator values (e.g. low = one SD below the mean; moderate = the sample’s mean; and high = one SD above the mean) and then estimating the effect of the focal predictor at those values (Hayes & Matthes, 2009). These analyses helped discover the specific role of each personal characteristic in the link between PMIEs and the psychiatric symptomatology of GSI/PTSS after deployment. All analyses were conducted with IBM SPSS software (V. 26). For legal reasons, supporting data for this study are unavailable.

3. Results

3.1. Prevalence of PMIEs, psychiatric symptoms, and PTSS after deployment

In this section, we calculated descriptive statistics and rates of exposure to PMIEs, PTSS, and psychiatric symptoms at T3. Regarding the PMIEs, 30.1% of the combatants endorsed at least one of the MIES-perpetration by oneself items, 33.6% endorsed at least one of the MIES-perpetration by others items, and 43.6% endorsed at least one of the MIES- betrayal items, at the ‘slightly agree’ or higher level (see Table 1 for full item endorsements). According to the PCL-5, 37 participants (11.1%) exceeded the 33-cut-off score anchored to stressful experiences in military service according to the DSM-5 (APA, 2013) criteria. The GSI, as defined by the BSI cut-off point, has been set to equal to or higher than the sum score of .71, according to the norms of Israeli males (Gilbar & Ben-Zur, 2002). Accordingly, the prevalence of current GSI was 26.9% (n = 90).

Table 1. Pearson correlation coefficients between the study variables (n = 322).

	1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. Cognitive index (T1)	–
2. Performance Prediction (T1)	−.11*	–
3. NLE- pre-enlistment (T2)	−.10*	.07	–
4. GSI - pre-enlistment (T2)	−.01	−.05	.11*	–
5. PTSS- pre-enlistment (T2)	−.01	−.03	.13*	.71***	–
6. ERQ (T2)	.05	.06	−.03	.11*	.10*	–
7. BIS (T2)	−.19***	.01	.11*	.60***	.34**	.01	–
8. AGQ (T2)	−.11*	.00	.25***	.45***	.49***	.01	.51***	–
9. COMBAT EXPOSURE (T3)	−.02	−.03	.10	.03	.12*	.08	−.01	.06	–
10. MIES-SELF(T3)	.12*	−.07	−.06	.12*	.11*	.03	−.02	.03	.17**	–
11. MIES-OTHER(T3)	.02	−.07	−.11*	.11*	.14*	.06	.01	.02	.11*	.61***	–
12. MIES-BETRAYAL (T3)	.01	.02	−.03	.23***	.24***	.01	.01	.22**	.19**	.51***	.41***	–
13. GSI-post deployment (T3)	−.02	−.07	−.00	.51**	.43***	.03	.21***	.38***	.21***	.37***	.30***	.41***	–
14. PTSS- post deployment (T3)	−.05	−.05	−.00	.40***	.41***	.03	.15**	.31***	.27***	.41***	.34***	.41***	.68***	–
M	57.31	31.42	2.54	.79	16.85	43.46	72.86	2.45	1.84	7.49	4.57	6.84	.56	11.55
SD	16.80	19.85	2.93	.74	16.52	9.07	14.26	.66	2.31	4.88	2.79	3.86	.67	13.62

Note. *p < .05. **p < .01. ***p < .001; NLE = Sum of negative life events before military service; GSI = Global severity index of the brief symptom inventory (BSI); PTSS = Posttraumatic stress symptoms for event(s); ERQ = Emotional regulation questionnaire; BIS = Barratt Impulsivity scale; AGQ = aggression questionnaire; MIES = Moral injury exposure scale.

3.2. Pre-enlistment personal characteristics and pre-deployment personality factors as contributors to PMIEs at T3

3.2.1. Preliminary analysis

To test to what extent the pre-enlistment personal characteristics and personality factors of emotional regulation, impulsivity, and aggression at T2 could account for a high proportion of PMIEs after deployment among soldiers, correlations between the study variables were calculated. The means, standard deviations, ranges, and intercorrelations for all variables are presented in Table 1. The matrix shows that the cognitive index was positively related to exposure to PMIE-Self (T3) and that aggression (T2) was positively related to exposure to PMIE-Betrayal (T3). Moreover, combat exposure (T3) was positively correlated with all three PMIE dimensions (T3).

3.2.1. Hierarchical regression analyses

To determine if the independent variables can predict PMIEs of self, other, and betrayal among soldiers beyond pre-enlistment psychiatric symptomatology indicators, we conducted three multiple hierarchical regression analyses (Cohen et al., 2003), with PMIE dimensions as the dependent variables. The three regression equation designs comprised four steps, each containing the same set of variables. Table 2 presents the contribution of each variable in the enter-step to the regression. The total set of variables in the final model explained 7.5% for PMIE-Self, 4.6% for PMIE-Other and 11.0% for PMIE-Betrayal. As seen in Table 3, we found that GSI before enlistment contributed positively to PMIE-Self and PMIE-Betrayal. Combat exposure, entered in the final step, positively correlated with all three PMIE dimensions. Personal characteristics at MIC (T1) and personality factors of emotional regulation, impulsivity, and aggression (T2) did not significantly contribute to PMIEs (T3).

Table 2. Three hierarchical regressions predicting MIES dimensions by study variables (n = 322).

Step	Predictors which enter in each step:	MIES-Self (T3)				MIES-Other (T3)				MIES-Betrayal (T3)
		B	t	Beta	ΔR^2	B	t	Beta	ΔR^2	B	t	Beta	ΔR^2
1	NLE- pre-enlistment (T2)	−.08	−1.50	−.08	.01	−.11	−1.97	−.11*	.02*	−.00	.10	−.00	.05***
	GSI - pre-enlistment (T2)	.14	−2.06	.12*		.10	1.92	.11		.22	4.91	.23***
2	Cognitive index (T1)	.11	1.83	.11	.02*	.00	.11	.00	.00	.02	.37	.02	.00
	Performance Prediction (T1)	−.10	−1.39	−.08		.07	−.97	−.06		.02	.26	.01
3	ERQ (T2)	−.02	−.44	−.03	.00	.03	.53	.03	.00	−.02	−.51	−.02	.01
	BIS (T2)	−.04	-−.52	−.02		−.09	−1.30	−.09		−.06	−.56	−.06
	AGQ (T2)	−.02	−.22	−.04		.01	.165	.01		.15	1.50	.16
4	COMBAT EXPOSURE (T3)	.20	3.35	.19**	.04***	.13	2.20	.13*	.02*	.22	3.86	.21***	.04***

Note. * p < .05. ** p < .01. *** p < .001; NLE = Sum of negative life events before military service; GSI = Global severity index of the brief symptom inventory (BSI); PTSS = Posttraumatic stress symptoms for event(s); ERQ = Emotional regulation questionnaire; BIS = Barratt Impulsivity scale; AGQ = aggression questionnaire; MIES = Moral injury exposure scale.

Table 3. Two hierarchical regressions predicting GSI-T3 and PTSS-T3 by the study variables (n = 322).

Step	Predictors	GSI-T3				PTSS-T3
		B	t	Beta	ΔR^2	B	t	Beta	ΔR^2
1	NLE- pre-enlistment (T2)	−.07	−1.40	−.07	.26***	−.06	−1.15	−.04	.17***
	GSI - pre-enlistment (T2)	.49	10.43	.52***		–	–	–
	PTSS- pre-enlistment (T2)	–	–	–		.38	8.06	.42***
2	Cognitive index (T1)	−.01	.17	−.01	.00	−.00	−.15	−.00	.00
	Performance Prediction (T1)	−.09	1.49	−.08		−.03	−.51	−.03
3	ERQ (T2)	−.03	−.60	−.03	.02*	−.03	−.67	−.04	.01
	BIS (T2)	−.10	−1.61	−.10		−.04	−.61	−.04
	AGQ (T2)	.17	2.57	.18*		.13	1.97	.15*
4	COMBAT EXPOSURE (T3)	.07	1.49	.07	.00	.25	4.91	.25***	.06***
5	MIES-SELF(T3)	.10	1.49	.10	.07***	.03	.50	.03	.06***
	MIES-OTHER(T3)	.05	.78	.05		.09	1.34	.09
	MIES-BETRAYAL (T3)	.18	3.25	.19**		.17	3.07	.18**
6	ERQ* MIES-BETRAYAL	−.18	−3.21	−.19**	.06***	−.19	−3.17	−.20**	.04*
	BIS* MIES-BETRAYAL	−.16	−2.22	−.16*		−.05	.07	−.05
	AGQ* MIES-BETRAYAL	.20	3.26	.22**		.05	.08	.05

Note. *p < .05. **p < .01. ***p < .001; NLE = Sum of negative life events before military service; GSI = Global severity index of the brief symptom inventory (BSI); PTSS = Posttraumatic stress symptoms for event(s); ERQ = Emotional regulation questionnaire; BIS = Barratt Impulsivity scale; AGQ = aggression questionnaire; MIES = Moral injury exposure scale.

3.3. Personal characteristics and personality factors before deployment and PMIEs as contributors to psychiatric symptomatology at T3

3.3.1. Preliminary analysis

To test to what extent a sense of pre-enlistment personal characteristics from the MIC and pre-deployment personality factors of emotional regulation, impulsivity, and aggression at T2 and PMIEs could account for psychiatric symptomatology after deployment among soldiers, we calculated the correlations between the study variables. The means, standard deviations, ranges, and intercorrelations for all variables are presented in Table 1. The matrix shows that pre-enlistment psychiatric symptoms and PTSS were all positively correlated with BSI and PTSS after deployment (T3). Higher levels of impulsivity and aggression before deployment (T2) were correlated with higher levels of BSI and PTSS after deployment (T3). In the same vein, all PMIE dimensions, as well as combat exposure, were also positively correlated with higher levels of BSI and PTSS after deployment (T3).

3.3.2. Hierarchical regression analyses

To determine if the independent variables can predict psychiatric symptomatology among soldiers beyond psychiatric symptomatology indicators prior to enlistment, we conducted two multiple hierarchical regression analyses (Cohen et al. 2003), with BSI-T3 and PTSS-T3 as the dependent variables. The two regression equation designs comprised six steps, each containing the same set of variables. Table 3 presents the contribution of each variable when entered into the regression. The total set of variables in the final model explained 42.1% of GSI-T3, F_(20,289)= 10.49, p < .001, and 35.0% for PTSS-T3, F_(20,289)= 7.77, p < .001.

Regarding GSI-T3, we found that GSI before enlistment contributed positively to GSI after deployment, adding 26% to the explained variance [F_{change(2,307)}= 54.47, p < .000]. Pre-enlistment personality factors entered in step 3 added 2% to the explained variance [F_{change(3,302)}= 2.65, p < .05], with aggression positively contributing to GSI-T3. In the fifth step, PMIEs dimensions added 7% to the explained variance [F_{change(3,298)}= 11.28, p < .001], with PMIE-Betrayal positively contributing to GSI-T3. In the final step, the nine interactions of risk factors and PMIE dimensions were entered, and three were found significant. Due to presentation issues, only the significant interactions are presented in Table 3 (the entire table, including the non-significant interactions, is presented in the Online Supplementary Appendix 1). This final step added 6% to the explained variance, F_{change(9,289)}= 3.17, p < .001, with the interactions of PMIE-Betrayal with emotional regulation, impulsivity, and aggression significantly contributing to GSI-T3.

Regarding PTSS-T3, we found that PTSS before enlistment contributed positively to PTSS after deployment, adding 17% to the explained variance, [F_{change(2,306)}= 32.25, p < .000]. Pre-enlistment personality factors entered in step 3 added 1% to the explained variance [F_{change(3,302)}= 1.51, Not significant], with aggression contributing positively to PTSS-T3. In the fourth step, combat exposure contributed positively to PTSS-T3, adding 6% to the explained variance [F_{change(1,301)}= 24.16, p < .000]. In the fifth step, PMIE dimensions added 6% to the explained variance [F_{change(3,298)}= 8.65, p < .001], with PMIE-Betrayal contributing positively to PTSS-T3. In the final step, the nine interactions of emotional regulation, impulsivity and aggression and three PMIE dimensions were entered, adding 4% to the explained variance [F_{change(9,289)}= 2.04, p < .05], with the interaction of PMIE-Betrayal with emotional regulation it was found to contribute significantly to PTSS-T3.

3.4. Moderation analyses

Following the hierarchical regression results, we conducted moderation analyses only for the significant interactions using PROCESS macro (Hayes, 2013). The probing of the interactions is presented in Figure 1a–d.

Figure 1. (a) The association between the MEIS-Betrayal and GSI-T3 as moderated by levels of emotion regulation (n = 322). Note. GSI = Global severity index of the brief symptom inventory (BSI); ERQ = Emotional regulation questionnaire; MIES = Moral injury exposure scale. (b) The association between the MEIS-Betrayal and GSI-T3 as moderated by levels of impulsivity (n = 322). Note. GSI = Global severity index of the brief symptom inventory (BSI); BIS = Barratt impulsivity scale; MIES = Moral injury exposure scale. (c) The association between the MEIS-Betrayal and GSI-T3 as moderated by levels of aggression (n = 322). Note. GSI = Global severity index of the brief symptom inventory (BSI); AGQ = aggression questionnaire; MIES = Moral injury exposure scale. (d) The association between the MEIS-Betrayal and PTSS-T3 after deployment as moderated by levels of emotional regulation (n = 322). Note. PTSS = Posttraumatic stress symptoms for event(s); ERQ = emotional regulation questionnaire; MIES = Moral injury exposure scale.

3.4.1. GSI-T3 as an outcome measure

Three moderation analyses of the significant interactions were conducted, all with GSI-T3 as the dependent variable and MIES-Betrayal as the independent variable. The moderators were the pre-enlistment personality factors of emotional regulation, impulsivity, and aggression. All other study variables were entered as covariates. As can be seen in Figure 1a, a significant interaction was found between MIES-Betrayal and emotional regulation, b = −0.91, SE = 0.03, 95% CI [−0.15, −0.03], t(287) = 2.98, p = .031. Probing the interaction revealed that for soldiers with low and moderate levels of emotional regulation, MIES-Betrayal was significantly associated with higher levels of GSI-T3: for low ERQ, b = 0.20, SE = 0.04, 95% CI [0.11, 0.29], t(287) = 4.39, p < .001; for moderate ERQ, b = 0.13, SE = 0.03, 95% CI [0.04, 0.19], t(287) = 3.24, p < .001. However, when emotional regulation levels were high, MIES-Betrayal did not predict GSI-T3 levels, b = 0.02, SE = 0.05, 95% CI [−0.71, 0.12], t(287) = 0.43, p = .792.

Figure 1b shows a significant interaction between MIES-Betrayal and impulsivity, b = −0.11, SE = 0.04, 95% CI [−0.20, −0.01], t(287) = −1.20, p = .027. Probing the interaction revealed that for soldiers with low and moderate levels of impulsivity, MIES-Betrayal was significantly associated with higher levels of GSI-T3: for low BIS, b = 0.21, SE = 0.05, 95% CI [0.09, 0.31], t(295) = 3.76, p < .001; for moderate BIS b = 0.09, SE = 0.03, 95% CI [0.02, 0.17], t(295) = 2.47, p = .012. However, when impulsivity levels were high, MIES-Betrayal did not predict GSI-T3 levels, b = 0.00, SE = 0.06, 95% CI [−0.12, 0.12], t(147) = −0.01, p = 991.

As seen in Figure 1c, we found a significant interaction between MIES-Betrayal and aggression, b = 0.82, SE = 0.25, 95% CI [0.02, 0.14], t(295) = 2.80, p = .05. Probing the interaction revealed that for soldiers with high and moderate levels of aggression, MIES-Betrayal was significantly associated with higher levels of GSI-T3: for high ARQ, b = 0.19, SE = 0.04, 95% CI [0.10, 0.27], t(295) = 4.19, p < .001, and for moderate ARQ, b = 0.09, SE = 0.04, 95% CI [0.04, 0.19], t(295) = 2.21, p = .027. However, when aggression levels were low, MIES-Betrayal did not predict GSI-T3 levels, b = 0.02, SE = 0.05, 95% CI [−0.71, 0.21], t(147) = 0.47, p = 638.

3.4.2. PTSS-T3 as an outcome measure

One moderation analysis was conducted with PTSS-T3 as the dependent variable, MIES-Betrayal as the independent variable and emotional regulation as the moderator. All other study variables were entered as covariates. As can be seen in Figure 1d, a significant interaction was found between MIES-Betrayal and aggression, b = −.13, SE = 0.04, 95% CI [−0.22, −0.04], t(295) = −2.88, p = .004. Probing the interaction revealed that for soldiers with low and moderate levels of emotional regulation, MIES-Betrayal was significantly associated with higher PTSS-T3: For low ERQ, b = 0.29, SE = 0.07, 95% CI [0.15, 0.43], t(295) = 4.21, p < .001; for moderate ERQ, b = 0.18, SE = 0.05, 95% CI [0.07, 0.29], t(295) = 3.22, p = .001. However, when emotional regulation levels were high, MIES-Betrayal did not predict PTSS-T3 levels, b = 0.03, SE = 0.07, 95% CI [−0.11, 0.18], t(147) = 0.48, p = 631.

4. Discussion

Along with the dangers in combat, modern military service exposes combatants to ambiguous situations within civilian and non-combatant surroundings, resulting in moral injuries with vast ranges of negative mental health outcomes (Griffin et al., 2019; Hall et al., 2022; Maguen et al., 2022). In this study, we aimed to shed light on the roles of pre-enlistment characteristics and pre-deployment personality factors that may facilitate PMIEs and moral injuries among combatants. We focused on psychiatric and posttraumatic stress symptoms in a two-year prospective study among IDF combatants, mainly deployed for peacekeeping missions in Israel's West Bank. To the best of our knowledge, this is the first study to examine the roles of such factors prospectively and their relation to PMIEs and MI-related psychopathology among active military Combatants.

A critical challenge in the study of MI is identifying transgressive acts and their harmful features that may facilitate psychopathological outcomes. To date, the most widely used PMIE assessment tool is the MIES (Nash et al., 2013). However, this measure might also constrain our ability to delineate the boundaries between specific transgressive acts, the subjective perception of participating in or witnessing them, and their possible deleterious ramifications (Griffin et al., 2019). Thus, although the outcome variables in the present study were PTSS and global distress, interpreting the findings should be made cautiously with regard to PMIE conflation of exposure and distress. Thus, for the remainder of this paper, we will use the term ‘subjective appraisal of PMIEs’ (i.e. subjective appraisals of causes and effects related to PMIEs), which better reflects the complexity of the MIES’s assessment of PMIEs.

The first question of this study related to the contribution of pre-enlistment characteristics and pre-deployment personality factors to subjective appraisal of PMIEs among combatants. Although we found that between 4.6% and 11.0% of the variance in subjective appraisals of PMIE dimensions was explained by the study variables, none of the pre-enlistment characteristics and pre-deployment personality factors significantly contributed to higher subjective appraisal of PMIE dimensions. While these are negative results, they may indicate that, at least to some extent, the subjective appraisal of PMIEs are relatively unrelated to personal or personality characteristics but rather to the military context and, specifically, to the extent of combat exposure (Frankfurt & Coady, 2021). In other words, it is probably not who you are that matters if you should encounter PMIEs, but the amount of potential moral injury situations you experience.

It is important to view these findings with caution in light of the characteristics of our combatant sample and, more specifically, the short time since deployment. Holowka et al. (2012) found that negative emotionality and high aggressiveness contributed to higher PMIEs among Vietnam War veterans almost 40 years post-deployment. Thus, personality factors may contribute to PMIEs only several years after the events. It may be that the chronological distance or older age from the actual events of PMIE may change perceptions of veterans and enable a deeper understanding of what happened (in terms of the transgression acts), and thus PMIEs may become more related to the participant’s personality (Zerach & Levi-Belz, 2022). Importantly, however, as our results suggest, the pre-enlistment psychiatric symptomatology significantly and positively contributes to higher levels of PMIE self and betrayal. Thus, combatants who struggled with psychiatric difficulties (e.g. depression) are more vulnerable to exposure to PMIEs. These results align with the conceptual model of Litz et al. (2009), which emphasised that neuroticism levels may facilitate the dissonance features of subjective appraisal of PMIEs, which result in depressive attributes, shame, and guilt.

The second and third questions of this study were related to the contributions of pre-enlistment characteristics, pre-deployment personality factors and subjective appraisals of PMIEs to PTSS and psychiatric symptomatology and the moderation role and pre-deployment personality factors in the associations between exposure to subjective appraisals of PMIEs, to PTSS and psychiatric symptomatology following deployment. Regarding subjective appraisals of PMIEs, our results indicate that all PMIEs dimensions were significantly and positively correlated with both GSI and PTSS post-deployment. However, only betrayal-based PMIEs contributed significantly to both GSI and PTSS, above and beyond pre-enlistment and pre-deployment characteristics, including pre-enlistment negative life events, psychiatric symptomology, and posttraumatic symptoms. These results are in line with the MI conceptual model (Litz, et al., 2009) and studies demonstrating the harmful effects of subjective appraisals of PMIEs among combatants (Koenig et al., 2019; Nazarov et al., 2018) and veterans (Maguen et al., 2022; Levi-Belz, Dichter et al., 2022). However, the current results are significantly important due to their prospective nature among a non-US sample of service members.

We found that only pre-deployment personality factors of aggression contributed to psychiatric symptomology and PTSS post-deployment. Interestingly, however, we found significant interactions between pre-deployment personality factors and PMIE-Betrayal on psychiatric symptomology: As the combatants are characterised by higher levels of aggression and lower levels of emotional regulation and impulsivity, the influence of PMIE-Betrayal on psychiatric symptomology was stronger (beyond the contribution of each factor alone and of pre-enlistment psychiatric symptomatology and negative life events). Similar results were found for emotional regulation regarding PTSS post-deployment.

Some of these results are not surprising. For example, the finding regarding the direct contribution of aggressiveness to PTSS and psychiatric symptoms in the present study is in line with other prospective studies which emphasised that high negative affect (DiGangi et al., 2013), pre-deployment hostility and anger (van Zuiden et al., 2011), are associated with more PTSD. Nevertheless, this is the first prospective study to also highlight the interactive effect of aggressiveness and emotional dysregulation with exposure to betrayal-based combat experiences on psychopathology following deployment.

Our results suggest that combatants who lack the ability to regulate harsh emotions or are characterised by higher aggressiveness may have difficulties in implementing mental processing, which may ease the inner moral conflicts they encounter following exposure to PMIEs of betrayal. As Litz et al. (2009) emphasised, they may continue to attribute their actions as global and stable and thus experience shame and guilt for not doing enough during PMIEs and in the face of commanders or authorities, which results in higher levels of distress and PTSS (Zerach & Levi-Belz, 2018). These notions are consistent with several studies which stressed that low emotional regulation and high aggressiveness might moderate the links between exposure to PMIEs and psychiatric symptoms (e.g. Forkus et al., 2021).

The results regarding the moderating effect of impulsivity are somewhat surprising, as scientific data view high impulsivity as a risk factor for psychopathology (Gvion et al., 2014; Netto et al., 2016). However, our results suggested that the association between PMIE-Betrayal and psychiatric symptoms is stronger at low levels of impulsivity (see Figure 1b). While this result requires more confirmation before being able to draw conclusions, it may be suggested that combatants with high impulsivity had a much greater ability to use self-forgiveness as a buffer against the negative outcomes of moral injury. Significantly, self-forgiveness has been recognised as an important protective factor in cases of MI (Litz et al., 2009; Levi-Belz, Dichter et al., 2022). In other words, if a combatant views himself as impulsive, he may forgive himself more readily after making regretful moral decisions, partly because he can interpret the situation as ‘I’m still a moral person, but my impulsiveness gets the better of me at times.’ However, a more calculated and settled combatant may find PMIEs to be more directly related to his moral decisions, perhaps resulting in shame, guilt, and moral injury outcomes (Hall et al., 2022). As no studies to date have examined this specific influence of impulsivity as related to moral decisions (in contrast to non-moral decisions and actions), this suggestion may remain only plausible until replications are conducted.

Some limitations of the present study should be outlined. First, although T1 measurements relied on semi-structural interviews, in T2–T3, we used self-report measures, which may suffer from various biases. Second, pre-enlistment PTSS and psychiatric symptomatology were measured only at T2, thus introducing a well-known range of biases caused by factors such as mood-dependent recall, forgetting, cathartic effect, and social desirability. Third, as this is the only study to include an Israeli active-duty combatants’ sample, cultural and military context differences, such as types of exposure and mandatory military service, should be acknowledged. Last, data were derived from several units within one brigade, which, although considered diverse and heterogeneous in terms of personal characteristics of combatants, its results should be validated within other combat units.

4.1. Conclusions and implications

In this study, we prospectively examined the moderation role of pre-deployment personality factors (at T2) on the subjective appraisal of PMIEs and psychopathological outcomes associations. Our results highlight that pre-deployment emotional regulation, impulsivity and aggressiveness levels should be assessed, screened, and identified among combatants, as they all facilitate psychiatric symptomology (and PTSS) after combatants have been exposed to PMIEs of betrayal. Considering that almost half of the combatants in our sample reported having experienced PMIE-Betrayal, such pre-assessment will facilitate identifying at-risk combatants and provide them with pre-deployment preparation regarding moral and ethical situations (Bloom et al., 2020). Even a psycho-educational protocol regarding this subject should be considered, as it may help diminish the negative consequences of MI found in our study. Moreover, the high percentage of combatants’ subjective appraisals of PMIEs and their deleterious effects also highlights the importance of strengthening the familiarity with MI among the chain of command as well as the medical and mental health personnel, and the combatants themselves. Together, such steps of pre-deployment assessment, identification and interventions are essential not only for operational mission success but also for the well-being of its participants.

Acknowledgments

We thank the late Yehuda Rubinovich, Mental Health Officer of the Kfir Brigade, for his contribution to the recruitment of the combatants to this project.

Disclosure statement

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

Data availability statement

The participants of this study did not give written consent for their data to be shared publicly, so due to the sensitive nature of the research supporting data is not available.

Additional information

Funding

The research reported in this publication was supported by the Israel Defense Forces (IDF) Medical Corps and Directorate of Defense Research & Development, Israeli Ministry of Defense (IMOD DDR&D), under [grant number: 1994/19].