(PS6-C73) Moral Injury Is Associated with Loneliness in US Veterans

Loneliness is associated with an increased risk of a variety of adverse psychological and physical health outcomes (Mullen et al., 2019; Park et al., 2020) such as depression (Straus et al., 2022; Teo et al., 2018) and suicidal ideation (Pietrzak et al., 2017; Teo et al. 2018). Veterans appear especially vulnerable to loneliness (Kuwert et al., 2014; Straus et al., 2022), which may put them at higher risk of such adverse outcomes. Both exposure to potentially traumatizing experiences (Kao et al., 2014; Liberatore-Maguire et al., 2022) and PTSD symptoms (Itzhaky et al., 2017; Solomon et al., 2015) are positively associated with loneliness. One path by which exposure to military stressors may increase loneliness is through moral injury (MI). MI is a syndrome of biopsychosocial outcomes that can arise after experiencing events that violate one’s deeply-held beliefs about right and wrong, otherwise known as potentially morally injurious events (PMIEs; Drescher et al., 2011; Jinkerson, 2016; Litz et al., 2009). MI can be disaggregated into shame-related (SR) symptoms, stemming from committing or failing to prevent a transgression, and trust-violation-related (TVR) symptoms, resulting from bearing witness to or being the victim of a transgression (Litz et al., 2022). Both the (expected or actual) social exclusion/rejection associated with SR symptoms and the loss of faith in humanity associated with TVR symptoms may lead to loneliness (Kvitsiani et al., 2023). In this secondary analysis of cross-sectional data collected from US Veterans (n = 265) in the Moral Injury Outcome Scale (MIOS) development project (Litz et al., 2022), we tested the hypothesis that, when controlling for PTSD symptoms, there would be an incremental, positive association between MI severity and reported loneliness. Using linear regression controlling for PTSD severity (F[2, 261] = 74.59, p < .001), we found that overall MI severity (measured by total score on the MIOS) was significantly associated with reports of loneliness (measured by total score on the UCLA Loneliness scale; Russell, 1996; β = 0.22, p = .003). We then ran two exploratory linear regressions with total scores on the SR and TVR subscales of the MIOS as independent variables. Overall, both models were significant (p < .001), however, TVR symptoms significantly predicted loneliness (β = 0.28, p < .001) while SR symptoms did not (p = .115). Our results suggest that MI is implicated in the etiology of loneliness in US Veterans and this relationship may be primarily driven by trust-violation-related MI.