Psychospiritual Developmental Risk Factors for Moral Injury

Abstract

There is increasing theoretical, clinical, and empirical support for the hypothesis that psychospiritual development, and more specifically, postconventional religious reasoning, may be related to moral injury. In this study, we assessed the contributions of exposure to potentially morally injurious events, posttraumatic stress symptoms, and psychospiritual development to moral injury symptoms in a sample of military veterans (N = 212). Psychospiritual development was measured as four dimensions, based on Wulff’s theory juxtaposing conventional vs. postconventional levels of religious reasoning, with decisions to be an adherent or a disaffiliate of faith. After controlling for exposure to potentially morally injurious events and severity of posttraumatic stress symptoms, veterans who were conventional disaffiliates reported higher scores on the Moral Injury Questionnaire than conventional adherents, postconventional adherents, or postconventional disaffiliates. We conclude that the role of psychospiritual development offers a theoretical approach to moral injury that invites collaboration between social scientists, philosophers, theologians, and medical professionals.

1. Introduction

The risk of trauma exposure, and the subsequent psychological reaction to that exposure, has played a role in the human experience throughout our history as a species. Within the past several decades, the clinical and therapeutic orientation toward diagnosis and treatment for posttraumatic stress disorder (PTSD) has shifted progressively from a sole focus on physical threat and fear-based anxiety to a more inclusive acknowledgement of psychological symptoms that affect trauma survivors (Friedman et al. 2011). Moral injury, which accompanies about one third of combat-related PTSD cases (Bryan et al. 2018; Frankfurt and Frazier 2016; Stein et al. 2012), involves the emotional, behavioral, and spiritual sequalae of exposure to events that challenge cherished values and beliefs (Litz et al. 2009). Symptoms of moral injury include guilt, shame, loss of meaning/purpose, loss of trust in self, others or Deity, depression, anxiety, anger, re-experiencing moral conflict, suicidal ideation/behavior, self-injury, social withdrawal, re-orienting of value systems, and impaired interpersonal functioning (Jinkerson 2016; Litz et al. 2009; Maguen et al. 2009; Steenkamp and Litz 2013; Wisco et al. 2017; Zimmermann et al. 2016). Furthermore, co-occurrence of PTSD and moral injury is associated with increased risk for suicide (Bryan et al. 2018). A growing number of studies indicate that a primary mechanism of impairment stemming from a moral injury is spiritual distress (Jinkerson 2016; Kopacz et al. 2018; Litz et al. 2009; Maguen et al. 2009; Steenkamp and Litz 2013; Wisco et al. 2017).

Spiritual distress can be a particularly salient factor for members and veterans of the Armed Forces exposed to morally ambiguous combat situations (Jinkerson 2016). Spiritual distress has been shown to be an important mediator between PTSD symptoms and moral injury (Evans et al. 2018). It may manifest as struggles with religious/spiritual communities, theological and moral principles, and one’s relating to a higher power (Exline et al. 2014). Mounting research around spirituality and trauma-related issues suggests a critical link between spiritual distress and clinical outcomes for many survivors (Bryan et al. 2018; Currier et al. 2014, 2015a; Harris et al. 2008, 2012, 2018). For example, individuals with higher levels of spiritual distress also typically have more severe PTSD symptoms (Bryan et al. 2018; Fontana and Rosenheck 2004; Harris et al. 2008, 2012) and a substantially longer course of PTSD symptoms (Fontana and Rosenheck 2004; Harris et al. 2012). Spiritual distress at baseline has also been shown to reduce response to PTSD treatment over time among veterans (Currier et al. 2015a). Emerging research is even showing that individuals who lose their faith in the context of trauma have far worse mental health outcomes than those who maintain their pre-existing religious faith (Ben-Ezra et al. 2010; Fontana and Rosenheck 2004; Ter Kuile and Ehring 2014).

Despite the fact that not every veteran identifies with a specific spiritual orientation, the subset whose understandings of morality are based in a spiritual identity require specific clinical attention to their spiritual distress in order to optimize treatment for their trauma-related symptoms (Harris et al. 2015). Spiritually integrated care is an emerging and significant treatment methodology in alleviating mental health challenges among veterans. More than 90% of veterans believe in a Higher Power and 89% identify as Christian (Fontana and Rosenheck 2004). Because spirituality can provide a primary lens for situational meaning-making, it is particularly relevant during the most stressful experiences in individual’s lives—experiences in which meaning may be difficult to find but highly important for autobiographical memory integration and long-term health (Park 2005). The way in which individuals appraise these experiences as consistent or inconsistent with their global meaning systems impacts the spiritual distress that they encounter in trying to resolve and integrate these experiences into their spiritual worldviews (Park 2005). The more inconsistent the potentially morally injurious experience is with their spiritual worldview, the more severe their spiritual distress may be if they cannot resolve the dissonance (Kopacz et al. 2018).

The majority of veterans who are exposed to morally complex situations encounter them during a critical window for psychospiritual development: early adulthood (Harris et al. 2015). Throughout this time, young adults reassess their spiritual worldviews as a part of normal maturation and spiritual development, ideally moving from conventional to postconventional spirituality (Harris et al. 2015). However, when young adults encounter morally conflictual events that do not align with their previously established spiritual worldview, they may face critical psychospiritual developmental challenges as they attempt to establish meaning from their experience and reorganize their spiritual frameworks. This process may be particularly distressing for individuals in considerably concrete or conventional psychospiritual development stages (Harris et al. 2015; Nash et al. 2013). Due to the specific challenges of present-day war-zone environments (such as children being weaponized by insurgents) and the typical life-stage of veterans experiencing it, approaching moral injury with an understanding of psychospiritual development is key to facilitating positive treatment outcomes (Harris et al. 2015).

Psychospiritual developmental models based on cognitive and moral developmental theories—such as Fowler’s stage theory of faith development—have the greatest empirical support to date (Fowler and Dell 2006; Harris and Leak 2013; Fowler 1981). Fowler’s theory, which draws on work by both Piaget and Kohlberg, was based on interviews conducted with wide ranging age and faith groups within the U.S. (Fowler 1981). In Fowler’s theory, the ability to critically evaluate religious ideas, or “postconventional spiritual functioning,” is one construct that marks higher levels of spiritual development in most, but not all religious subcultures (Harris and Leak 2013). Ricoeur (1970) and Wulff (1991, 1997) have developed a model of postconventional spiritual functioning in the modern world that allows for postconventional acceptance or rejection of a Higher Power construct; individuals whose spiritual functioning is postconventional likely interpret scripture in an abstract, or symbolic context, while individuals functioning at conventional levels will interpret scripture more literally (Duriez et al. 2005).

Postconventional functioning allows individuals to process morally ambiguous events, to critically examine spiritual questions, and to consider multiple social and contextual perspectives on spiritual concerns, while individuals who are functioning at conventional levels may either defer to, or completely reject, religious or social authorities, considering spiritual and moral questions concretely and categorically, rather than on a continuum (Duriez et al. 2005; Harris et al. 2015). Theoretically, postconventional perspective-taking capacity provides for greater resilience in the face of potentially morally injurious events. This study is designed to test the hypothesis that psychospiritual development may be associated with moral injury syndrome. Based on the literature presented above, it is reasonable to hypothesize that, after controlling for exposure to potentially morally injurious events and severity of PTSD, individuals who function at a conventional level of psychospiritual development and have disaffiliated from faith will report more severe symptoms of moral injury syndrome. While a longitudinal test of these relationships would be ideal, establishing a cross-sectional relationship between these variables is an important first step and primary goal of this study.

2. Materials and Methods

2.1. Procedures/Design

The sample was drawn from the database of a VA Health Care System in the Midwest. Letters were mailed to a random sampling of Veterans with a PTSD diagnosis enrolled in care in this VA Health Care system. The project was reviewed and approved by the facility’s Institutional Review Board, and a passive consent letter was included in the survey mailings. Veterans were informed that the study was voluntary and that they would receive a $10 gift card for completing the survey. Veterans were also given the option to opt out of future mailings or call the principal investigator (initials) with any questions. Data were coded and entered with discrepancy checks done before analysis. The discrepancy check was conducted by a physical review of the data by a second research assistant to confirm the original entries in Statistical Package for the Social Sciences, Version 26.

2.2. Measures

The Multiscale Measure for Postconventional Spiritual Functioning (MMPSF; Harris and Usset 2017) was designed to combine the psychometric strengths of the Fowler Religious Attitudes Scale-Revised and the Post-Critical Belief Scale (PBS), while addressing the limitations of both. This 55-item Likert-type instrument yields scores in four subscales; Conventional-Adherent (maintains conventional religious reasoning, adheres to faith), Conventional-Disaffiliate (maintains conventional religious reasoning, disaffiliates from faith), Postconventional-Adherent (maintains postconventional religious reasoning, adheres to faith) and Postconventional-Disaffiliate (maintains postconventional religious reasoning, disaffiliates from faith). Coefficient alphas were as follows: Conventional Adherent (0.89), Conventional Disaffiliate (0.93), Postconventional Adherent (0.79) and Postconventional Disaffiliate (0.81) (Harris and Usset 2017). Construct validity was supported by an anticipated pattern of correlations with the Fowler Religious Attitudes Scale-Revised (Harris and Leak 2013). Only the Conventional Adherent scale has correlated with validation of an item asking if the participant was in the same faith in which they were raised (Harris and Usset 2017).

The Morally Injurious Events Scale (MIES; Nash et al. 2013) is a 9-item Likert type scale assessing exposure to potentially morally injurious events (PMIEs). This measure captures the occurrence of PMIEs using a six-point scale in which 1 = Strongly agree to 6 = Strongly disagree. Items can yield two general subscales, with six items capturing Perceived Transgressions based on one’s commission or omission of certain acts/decision and three capturing Perceived Betrayals by military and/or non-military individuals (Nash et al. 2013). Cronbach’s alpha for the total scale in the initial validation sample was 0.90 (Nash et al. 2013). Correlations with measures of moral injury syndrome symptoms, such as the Expressions of Moral Injury Scale (Currier et al. 2018), have supported validity (Lancaster and Harris 2018).

The PTSD Checklist-5 (PCL-5; Weathers et al. 2013) is a 20-item, Likert-type instrument assessing distress over the past month associated with symptoms of PTSD from the four DSM-5 domains (i.e., reexperiencing [e.g., nightmares], behavioral avoidance [e.g., avoiding reminders of event], dysphoric emotion and cognition [e.g., self-blame,], hyper-arousal [e.g., insomnia]). Items were rated on a five-point scale, ranging from 0 = Not at all to 4 = Extremely, and summed for a total symptom severity score. Cronbach’s alpha was 0.96 (Bovin et al. 2016). Validity is supported via similar diagnostic classifications to those derived from the Clinician Administered PTSD Scale (Bovin et al. 2016).

The Moral Injury Questionnaire (Currier et al. 2015b) is a 20-item, Likert-type scale designed to assess exposure to potentially morally injurious events. The MIQ-M includes 19 items that cover a broad range of possible combat-related activities/circumstances that might violate one’s moral beliefs/values. Currier et al. (2015b) provided preliminary evidence for the validity (factorial, convergent, and incremental) and utility of the instrument for additional research and clinical work with military populations. Further, construct validity has been supported has been supported in correlations with other self-report moral injury measures in more recent work (Lancaster and Harris 2018). In the present study participants were instructed to endorse the frequency with which they experienced each of the MIEs in the context of their war-zone deployment(s) to Iraq and/or Afghanistan on a 4-point scale in which 1 = Never, 2 = Seldom, 3 = Sometimes, and 4 = Often. The coefficient alpha in this sample was 0.91.

3. Results

3.1. Demographics

The sample included 212 veterans who had received VA PTSD care (172 male, 40 female) that expressed an interest in participating in survey research relevant to PTSD. The average age was 57.86 years (SD = 14.79 years) and the average level of formal education was 14.14 years (SD = 2.31 years). Racial identities included Caucasian (88%), African American (5%), Latinx (2%) and Other (5%). Religious affiliations included Protestant Christian (41%), Catholic (24%), Agnostic (8%), Atheist (3%), Buddhist (3%), Jewish (2%), Muslim (2%), Hindu (1%), and not specified (16%). Ninety-six percent of the sample scored above the cut-off score of 33 for a probable PTSD diagnosis on the PCL-5.

3.2. Analysis

See Table 1 for means and standard deviations and Table 2 for intercorrelations among study variables. While cutoff scores have no clinical meaning, as the MMPSF, does not assess psychopathology, scores that are more than one deviation above the mean, which could be defined as high scores on each subscale, would be 13.25 for the Conventional Adherent scale, 12.00 for the Conventional Disaffiliate scale, 20.63 for the Postconventional Adherent scale, and 23.91 for the Postconventional Disaffiliate scale. Higher scores on the Conventional Adherent scale of the MMPSF were related to lower scores on the MIQ, and higher scores on the Conventional Disaffiliate and Postconventional Disaffiliate scales were related to higher scores on the MIES and MIQ. PTSD symptoms were moderately positively related to scores on the MIES and MIQ.

Building on these initial results, we next compared the relative contributions of conventional adherence, conventional disaffiliation, postconventional adherence, and postconventional disaffiliation, while controlling for the effects of exposure to potentially morally injurious events and PTSD symptoms, on MIQ scores using hierarchical multiple regression analysis. Because of potential multicollinearity between the Conventional Disaffiliate and Postconventional Disaffiliate subscales, separate regression analyses were run, testing the Conventional subscales in one analysis, and the Postconventional subscales in another. Scores on the MIQ were the outcome variable.

In the first multivariate analysis, exposure to potentially morally injurious events was entered in Step 1, PTSD symptoms were entered in Step 2, and conventional adherence/disaffiliation or postconventional adherence/disaffiliation were entered in Step 3. In the final model, exposure to potentially morally injurious events (β = 0.36, p < 0.001), PTSD symptoms (β = 0.26, p < 0.001) and Conventional Disaffiliate scores (β = 0.27, p < 0.001) significantly predicted scores on the MIQ. No evidence suggested a statistically significant association between conventional adherence (p = 0.107) and moral distress (see Table 2).

In the final step of the second multivariate analysis with the postconventional subscales, exposure to potentially morally injurious events (β = 0.37, p < 0.001), PTSD symptoms (β = 0.29, p < 0.001) significantly predicted scores on the MIQ. No evidence suggested a statistically significant association between postconventional adherence (p = 0.76) or postconventional disaffiliation (p = 0.68) and moral distress in this analysis (see Table 3).

4. Discussion

In this sample, those with conventional levels of religious functioning that disaffiliated from their faith/belief tradition experienced greater degrees of suffering as measured by the MIQ. There was no evidence to suggest that postconventional disaffiliates experienced the same degree of suffering as their conventional counterparts. We observed a negative bivariate association between conventional adherence and moral distress, which may indicate that conventional adherents are less likely to experience moral distress if they circumvent morally ambiguous situations as a result of concrete categorizations and thinking processes (e.g., service members involved in a military operation in which a civilian was incidentally harmed contends that they were “just doing a job.”). Still, postconventional levels of psychospiritual development may be a resource that helps people to endure stress and make meaning out of adverse and ambiguous stressful events. These findings support the psychospiritual developmental model of moral injury introduced by Harris et al. (2015). Namely, that those with conventional levels of religious or belief system functioning may be at greater risk or experience higher levels of suffering after experiencing potentially morally injurious events. Compared to conventional disaffiliates, postcoventional affiliates and disaffiliates may experience fewer symptoms of moral injury because they are more likely to make meaning of PMIEs through integration of multiple moral contexts. While trauma-related disaffiliation from faith has emerged in the literature as a risk factor for worse mental health outcomes (Ben-Ezra et al. 2010; Fontana and Rosenheck 2004; Ter Kuile and Ehring 2014), when individuals move to a postconventional level of religious functioning, this effect does not appear in this cross-sectional study.

These findings add to a growing body of literature that supports extending study and care of those exposed to Potentially Morally Injurious Events (PMIEs) beyond the traditional fear-based pathology that characterizes PTSD. For example, VA chaplains are known to support Veterans with a high prevalence and high perceived intensity of PMIEs, relative to other military and Veteran groups (Kopacz et al. 2019). Accounting for both the content and way in which individuals hold their belief system or worldview (psychospiritual development) provides relevant insight into the understanding of how people from different religions, belief systems, races, cultures, ethnicities, and genders experience potentially morally injurious events. A need exists for more cross cultural studies of moral injury, especially given that an individual’s moral beliefs are oft internalized from group-based, societal, and cultural ethics that differ widely across people groups. Finally, moral injury research has largely been conducted in a siloed, discipline specific approach. Psychospiritual development offers a theoretical approach to moral injury that invites collaboration between social scientists, philosophers, theologians, and medical professionals.

Several interventions for moral injury have been developed over the past decade such as Acceptance and Commitment Therapy for Moral Injury (Farnsworth et al. 2017), Adaptive Disclosure (Litz et al. 2015), Building Spiritual Strength (Harris et al. 2018), Impact of Killing in War (Maguen et al. 2009) Trauma Informed Guilt Reduction (Norman et al. 2019), and Mental Health Clinician and Community Clergy Collaboration (Pyne et al. 2019). Adaptions have also been made to Prolonged Exposure (PE) and Cognitive Processing Therapy (CPT) to better account for PMIEs (Smith et al. 2013; Wachen et al. 2017). The relationship between Conventional Disaffiliates and higher MIQ scores suggests a need for therapists and chaplains leading these interventions to account for both psychospiritual development and affiliation or disaffiliation with a belief system community. More broadly, these findings support the need for facilitators of moral injury interventions to account for the meaning of an individual’s religious/spiritual values in their healing process.

5. Conclusions

5.1. Future Research Directions

All participants of this study were veterans, which limits generalizability. However, the onset of the COVID-19 pandemic has increased the interest and study of moral injury in healthcare worker and first responder populations. Further research on the measurement and impact of moral injury in healthcare workers and first responders will be crucial to understand how PMIEs are experienced in those populations (Maguen and Price 2020). A major barrier to this work is the absence of a measure to assess exposures and associated morally injurious outcomes in populations other than military service members and veterans. Furthermore, few studies examine what values veterans hold that they believe were violated by perceived transgressions. Research is needed to understand the relationship between psychospiritual development and specific value orientations (Zimmermann et al. 2016) or moral foundations (Graham et al. 2013). For instance, when asked to describe the types of military experiences that they might perceive to be morally ambiguous (Schorr et al. 2018), veterans often describe events where they must prioritize one core value (e.g., care for a vulnerable person such as an armed child) over another core value (e.g., loyalty to one’s fellow service members by eliminating potential risks).

Next, given the possibility of a longitudinal relationship between psychospiritual development and moral injury, further longitudinal studies are needed. For example, studies provide evidence of a religious residue effect, defined as cognitive, emotional, and behavioral ways that formerly religious people differ from people who identify as never or currently religious (Van Tongeren et al. 2020). Religious residue might contribute to moral injury, especially among Veterans who make abrupt and impassioned decisions to disaffiliate following exposure to military-related potentially morally injurious events. A caveat is that studies are needed that are both longitudinal and capture critical developmental events, such as the window in time during which a veteran makes a decision to disaffiliate perhaps corresponding with a combat deployment or in the first few years following separation from service. Additional research is also needed to examine how psychospiritual development might contribute to moral resilience. Future research is therefore indicated to better understand the relationship between psychospiritual development and spiritual fortitude (Van Tongeren et al. 2019), especially to the extent that spiritual fortitude aids in meaning making and well-being following exposure to traumatic events (McElroy-Heltzel et al. 2018).

5.2. Limitations

Given the cross-sectional nature of this study, causality or temporal associations between the variables cannot be determined. The MIQ includes questions measuring exposure to PMIEs and items that measure experiences of moral injury syndrome. Measurement development in moral injury is ongoing in the literature and will likely offer additional options for measuring exposure vs symptoms in the future. The present study’s sample was limited in diversity in gender, age, geographic location, and religious belief systems. Limitations were more pronounced in racial, gender, and ethnic diversity. Because the sample was drawn from those already seeking care at the VA, selection bias may have played a factor in sampling. For example, this sample possibly represents an over- or under-representation of highly distressed veterans such that the results possibly have limited generalizability outside of the veteran/service member populations. These findings also do not speak to the reasons Conventional/Postconventional Disaffiliates have withdrawn from their faith/belief system communities. Yet another concern is that we do not know the extent to which Conventional/Postconventional development is stable; further research is needed to determine if those who respond as postconventional would do so continuously over time. Nevertheless, there are important implications for communities of faith across the belief system spectrum raised by these findings.

Author Contributions

Conceptualization, T.J.U., B.J.G., J.M.C., M.S.K. and J.I.H.; Data curation, T.J.U., J.H.W. and J.I.H.; Formal analysis, J.I.H.; Funding acquisition, J.I.H.; Investigation, T.J.U. and J.I.H.; Methodology, T.J.U. and J.I.H.; Project administration, T.J.U., E.G. and J.I.H.; Resources, T.J.U. and J.I.H.; Software, J.I.H.; Supervision, T.J.U. and J.I.H.; Validation, T.J.U., J.H.W.and J.I.H.; Visualization, J.I.H.; Writing–original draft, T.J.U., E.G., B.J.G., J.M.C., M.S.K., J.H.W. and J.I.H.; Writing–review & editing, T.J.U., E.G., B.J.G., J.M.C., M.S.K., J.H.W. and J.I.H. All authors have read and agreed to the published version of the manuscript.

Funding

This study was funded by a grant from the John and Maria Laffin Trust.

Conflicts of Interest

The authors declare no conflict of interest.

Tables

Table 1

Means, Standard Deviations and Intercorrelations for Major Variables.

Variable	Mean	SD	CA	CD	PA	PD	MIES	PCL	MIQ
MMPSF: CA	21.02	7.77	0.89
MMPSF: CD	8.48	3.52	−0.38 **	0.93
MMPSF: PA	15.87	4.76	−0.04	−0.06	0.79
MMPSF: PD	16.59	7.32	−0.60 **	0.61 **	0.11	0.81
MIES	34.71	11.44	−0.08	0.15 *	0.07	0.09	0.90
PCL	58.05	14.29	−0.26 **	0.09	0.01	0.19 *	0.40 **	0.96
MIQ	49.84	18.29	−0.19 *	0.20 *	0.04	0.28 **	0.48 **	0.41 **	0.91

SD: Standard Deviation; CA: Conventional Adherent; CD: Conventional Disaffiliate; PA: Postconventional Adherent; PD: Postconventional Disaffiliate; MIES: Morally Injurious Events Scale; PCL: PTSD Checklist-5; MIQ: Moral Injury Questionnaire; * p less than or equal to 0.05; ** p less than or equal to 0.001 Coefficient alphas are on the diagonal for the correlation matrix.

Table 2

Exposure to Potentially Morally Injurious Events, PTSD Symptoms, and MMPSF conventional subscales as predictors of Moral Injury Questionnaire Score.

Step/Variable	B (p)	F (p)	Adj. R2	R2Δ	FΔ (p)
Step 1		44.50 (<0.001)	0.24
MIES	0.80 (<0.001)
Step 2		29.59 (<0.001)	0.29	0.06	11.37 (0.001)
MIES	0.64 (<0.001)
PCL-5	0.34 (0.001)
Step 3		18.33 (<0.001)	0.33	0.05	5.24 (0.006)
MIES	0.60 (<0.001)
PCL-5	0.34 (<0.001)
Conventional Adherent	0.24 (0.107)
Conventional Disaffiliate	0.43 (<0.002)

MIES = Morally Injurious Events Scale; PCL-5 = PTSD Checklist-5.

Table 3

Exposure to Potentially Morally Injurious Events, PTSD Symptoms, and MMPSF postconventional subscales as predictors of Moral Injury Questionnaire Score.

Step/Variable	B (p)	F (p)	Adj. R2	R2Δ	FΔ (p)
Step 1		38.57 (<0.001)	0.21
MIES	0.79 (<0.001)
Step 2		28.51 (<0.001)	0.29	0.08	14.66 (<0.001)
MIES	0.63 (<0.001)
PCL-5	0.37 (0.001)
Step 3		14.11 (<0.001)	0.27	<0.01	0.09 (0.92)
MIES	0.62 (<0.001)
PCL-5	0.37 (<0.001)
Postconventional Adherent	−0.06 (0.76)
Postconventional Disaffiliate	0.12 (0.68)

MIES = Morally Injurious Events Scale; PCL-5 = PTSD Checklist-5.