Discussion
The present study assessed program feasibility, satisfaction, and changes from pre- to post-intervention in moral injury outcomes among recent-era veterans who participated in a mindfulness versus educational support program for moral injury. Both programs were well received, particularly for the MMMI condition in which participants reported significantly more use of what they had learned, perceived more confidence in using what they had learned, and reported greater return on investment. During MMMI sessions, participants had a chance to practice mindfulness exercises which might have contributed to higher use and confidence scores. Also, results of our initial testing found MMMI treatment had a larger impact in decreasing moral injury symptoms than the ES treatment. Taken together, we found effect sizes consistent with change in moral injury symptoms within the time frame examined in this study. In general, effect sizes for condition x time effects were in the medium to large range, whereas after collapsing across conditions, our findings showed mostly large effect sizes (Richardson, 2011) on moral injury from pre- to post-treatment.
More specifically, when we examined pre-post changes by condition on effects sizes, we found four interactions for the treatment conditions over time. As compared to the ES condition, participants in the MMMI condition reported greater reductions in total MIOS scores, shame-related experiences due to moral injury, and impaired functioning due to moral injury on the MIOS domain (Litz et al., 2022). Results from the EMIS-M other-directed domain (Currier et al., 2018) showed veterans in the MMMI condition had reductions in moral injury symptoms associated with betrayal (e.g., betrayal, feelings of revenge, hostility toward authority). Shame-related moral injury (e.g., guilt, shame, self-condemnation) is theorized to result from acts for which the veteran experiences personal responsibility (e.g., injuring a non-combatant) or accidental, inappropriate, or prohibited behaviors the veteran witnessed but did not counter (e.g., watching peers treat civilians with disrespect but doing anything; Litz et al., 2009). Shame-related moral injury had been examined in Marines or soldiers who were directly attached to a combat arms unit (e.g., infantry, artillery) (e.g., Drescher et al., 2011; Flispe Vargas et al., 2013). We did not restrict our sample to veterans who were directly attached to a combat arms unit.
The condition x time effect for shame-related moral injury may reflect that sessions in the MMMI program focused on helping veterans use mindfulness to cultivate and practice extending compassion and acceptance toward themselves. Service members often have high standards regarding self-sacrifice, duty, and honor, which may increase guilt and shame from behaviors they perceive as unethical. In fact, a purported hallmark of moral injury entails difficulty with self-forgiveness for perceived or actual wrongdoing. This finding is important as veterans often have difficulty extending compassion and acceptance toward themselves particularly for events they perceive as their responsibility (see Purcell et al., 2018 for a discussion). Further, self-condemnation, self-harm, and self-sabotage are features of moral injury in which some veterans feel the need to punish themselves for self-perpetrated acts that took place in combat (Litz et al., 2009). Researchers have argued that fostering self-empathy and forgiveness are key to healing from moral injury (Evans et al., 2020; ter Heide, 2020).
We also found changes in other-directed (i.e., betrayal-based) moral injury as measured by the EMIS-M (Currier et al., 2018). MMMI focused on cultivating and practicing acts that would help veterans extend compassion and empathy toward others for actions that were transgressed against them. The distinction between self-directed (shame-related moral injury) and other-directed (betrayal-based) moral injury is important. Much of the existing literature has focused on self-directed moral injury, which stems from commission or omission of violence (e.g., Purcell et a., 2018). However, there is growing evidence that moral injury also comes from victimization, such as in the case of military sexual assault (Hamrick et al., 2022; Maguen et al., 2022) or acts for which no one is to blame (Fleming, 2021) and the moral injury symptoms may differ as a function of the type of events experienced (Currier et al., 2018).
After collapsing across conditions, significant pre-post effect size changes were found on all moral injury domains except the MP scale (Fleming, 2021). The lack of significant changes from pre-to post on the MP (Fleming, 2021) may reflect that most moral injury scales (e.g., EMIS-M, MISS, MIOS) focus on effects, symptoms, and moral emotions associated with moral injury (e.g., guilt, anger), whereas the MP scale assesses core views that are discrepant and disrupt underlying moral assumptions. These beliefs are not necessarily responses to acts of perpetration, omission, or betrayal (i.e., “I often think that life is absurd since my experience in the military”).