The association between substance use disorders and mortality among a cohort of Veterans with posttraumatic stress disorder: Variation by age cohort and mortality type

Abstract

BackgroundPrior studies of Veterans have linked posttraumatic stress disorder (PTSD) with an increased risk of mortality. Other studies of Veterans have found that substance use disorders (SUDs) are associated with an excess risk of mortality among those with psychiatric disorders. It is not known whether having an SUD increases the risk of mortality among Veterans with PTSD, and whether the association differs by mortality type or varies by age cohort. MethodsA cohort of patients who received Veterans Health Administration services during fiscal year (FY) 2004 and diagnosed with PTSD (n=272,509) were followed from FY 2005 through FY 2007 for the main outcomes of mortality and cause of death. ResultsSUD was positively associated with mortality during follow-up (adjusted hazards ratio: 1.70; 95% confidence interval: 1.64, 1.77). SUD was a stronger predictor of non-injury-related mortality for the <45 years group compared with the 45–64 or ≥65 group. SUD predicted injury-related mortality for all age groups. ConclusionsAmong Veterans with PTSD, the association between SUD and mortality was most pronounced for the youngest age group, which included Iraq/Afghanistan Veterans. For older age groups, which included Vietnam-era Veterans, SUD was a greater predictor of injury-related mortality. The findings could be useful for identifying PTSD patients at excess risk of mortality.