Michelle L Kelley, Jeffrey M Gabelmann, Megan Strowger, John Hearton, Folly Folivi, Adrian J Bravo, Jinjoo Noh, Kristin Kuskye, William Haber, Adam P McGuire
5 min
This study investigated the unique associations between four psychosocial factors—combat exposure, moral injury, PTSD symptoms, and pain—and various forms of substance use among U.S. military personnel. While previous research has linked these factors to substance use, this study sought to determine which specific factors uniquely predict the misuse of prescription opiates, prescription sedatives, cannabis, and hazardous alcohol consumption.
The researchers conducted a secondary analysis of baseline data from 238 U.S. military personnel who had deployed at least once. Participants completed standardized assessments for combat exposure, moral injury, PTSD symptoms, and pain. Substance use was measured through self-reports of past-week misuse of prescription opiates and sedatives, cannabis use, and hazardous alcohol consumption (using the AUDIT-C clinical cut-off). The team used multivariable multinomial logistic regression to isolate the unique contribution of each psychosocial factor to these different substance use outcomes.
The study found high rates of substance use within the sample: 21.0% reported prescription opiate misuse, 25.6% reported prescription sedative misuse, 14.7% reported cannabis use, and 46.2% met the criteria for hazardous alcohol use. The regression analyses revealed distinct patterns:
These findings suggest that substance use among military personnel is not a monolithic issue but is driven by different underlying psychological experiences. By identifying that moral injury and combat exposure have distinct impacts on prescription misuse compared to the role of PTSD in cannabis use, clinicians can better tailor treatment plans. The results highlight the necessity of screening for moral injury alongside traditional assessments like PTSD and pain to provide more effective, targeted care for veterans.
Background Prior research suggests that military personnel endorse higher rates of prescription drug misuse, cannabis, and heavy alcohol use than civilians. Factors related to substance use may differ for military personnel compared to civilians. In the present study, we examined whether combat exposure, moral injury, posttraumatic stress disorder (PTSD), and pain were associated with military personnel's self-reports of misuse of prescription opiates, prescription sedatives, both prescription opiates and sedatives, cannabis use, and hazardous alcohol consumption. Method Participants were a community sample of 238 U.S. military personnel who had deployed one or more times (71.0% males; M = 33.3 years; SD = 3.2). Results In our sample, rates of past week misuse were as follows: 21.0% prescription opiates, 25.6% prescription sedatives, 16.4% both prescription opiates and sedative medications. With respect to cannabis use and alcohol consumption, 14.7% reported past-week cannabis use and 46.2% participants reported hazardous alcohol consumption above suggested clinical cut-offs. In multivariable multinomial logistic regression analyses, combat exposure and moral injury were uniquely associated with a greater likelihood of misusing prescription opiates, sedatives, and both opiates and sedatives versus no misuse. Higher PTSD symptoms scores were uniquely associated with past week cannabis use versus no use. Further, greater combat exposure was uniquely associated with a greater likelihood of engaging in hazardous alcohol use. Conclusions These results suggest that distinct psychosocial factors may differentially impact substance use among military personnel. Findings indicate the importance of assessing combat exposure, moral injury, PTSD, and pain to better understand substance use and treatment of military personnel.
Sam: Exactly. And what survives that filter is telling. When all four are in the model together, PTSD and chronic pain lose their predictive power for prescription misuse. Moral injury and combat exposure retain significance. That's the load-bearing finding — not just that moral injury correlates with misuse, but that it does so independently, after the variance it shares with PTSD is accounted for.
Alex: Which also reframes the chronic pain result. Pain looked like a driver in isolation, but once you partial out the trauma constructs, it no longer predicts misuse independently.
Sam: [deliberate] Right — and that's worth flagging, because chronic pain is often the stated clinical rationale for opiate prescriptions in veteran populations. The model suggests pain may be a surface-level explanation, while the underlying driver is moral injury. Whether that's because moral injury amplifies pain perception, or because veterans with moral injury are more likely to seek pharmaceutical relief for psychological distress, the cross-sectional design can't tell us.
Alex: Which brings us to the limitations. How much weight can these findings actually bear? [[RP_SECTION:clinical-implications-and-limitations|Clinical Implications and Limitations]]
Sam: [measured, honest] The cross-sectional design is the primary constraint — no causal inference is warranted. The sample is 238 participants drawn from a daily diary study, so this is a secondary analysis on a relatively modest N. Self-report data on substance misuse carries underreporting risk, particularly for alcohol, even under anonymous conditions. And the model identifies the signal, but it doesn't explain the mechanism. Why does moral injury specifically map onto opiates and sedatives rather than cannabis? The paper doesn't resolve that.
Alex: [reflective] That's the gap a follow-up would need to close. You'd want longitudinal data, ideally with measures of the specific moral injury content — whether it's perpetration-based, betrayal-based — to see if substance choice tracks the injury type more granularly.
Sam: [calm, considered] That's the natural next step. What the paper does establish is that collapsing these constructs under a broad "trauma" umbrella obscures clinically meaningful variation. Moral injury isn't just a subtype of PTSD — it's a distinct predictor with a distinct behavioral signature. For clinical screening, that argues for moving beyond generic trauma assessment toward targeted moral injury measures, particularly when prescription misuse is part of the presenting picture.
Alex: [quiet conviction] It's a modest sample with real design constraints, but the signal is specific enough to warrant replication at scale. The argument that substance choice reflects the character of the underlying distress — not just its severity — is the kind of hypothesis that could meaningfully reshape how we triage veterans in clinical settings. Thanks for listening to ResearchPod.