A common assumption is that repression of traumatic memories is harmful to health. To assess this, we examined all-cause mortality among a national random sample of 4462 male US Army veterans evaluated in 1985 and followed up in 2000. Our hypothesis was that repression on the Welsh R scale would be associated with increased future mortality. We also expected to find a repression × post-traumatic stress disorder (PTSD) interaction effect. Multivariate Cox regression results for all veterans and for theater veterans (Vietnam service) and era veterans (no Vietnam service) separately, revealed that while PTSD was significant in all models, no main or interaction effect was found for repression. In addition, for era veterans, higher repression symptoms were protective for future mortality (HR = 0.95, p = 0.03). For hostility symptoms, although no interaction effect was found by PTSD, a positive main effect was detected for hostility, but only for theater veterans (HR = 1.04, p = 0.034). Disease-specific results were nonsignificant. Similar to a recent study, we also found that repression symptoms were negatively correlated with PTSD symptoms (r = −0.109, p < 0.001), suggesting repression might be protective. Our study found no evidence that repression had an adverse health impact on men exposed to psychological trauma.