Trends in unintentional injury death among post-9/11 Army Veterans who do and do not use Veteran Health Administration services

BACKGROUND: It is unknown if unintentional injury death rates vary among Army Veterans who do and do not use the Veterans Health Administration (VHA). We estimated, and compared by VHA-use group, age-adjusted average annual unintentional injury death rates and time-dependent hazard rates (HRs) after separation. METHODS: Data were from the Substance Use and Psychological Injury Combat Study, a longitudinal cohort of Army soldiers returning from an Afghanistan/Iraq deployment in FYs 2008-2014 using administrative data from the Military Health System, Veterans Health Administration and National Death Index. We selected soldiers who separated from active duty service by 9/30/2017 (n = 649,811). The exposure was time since separation, moderated by VHA-use group. The outcome was unintentional injury death (all, motor vehicle crash [MVC], overdose) through 2018. RESULTS: The HR for all unintentional injury and MVC death for 0-2 years post-separation were higher for the no VHA-use group than the VHA-use group (e.g., MVC, year 0-1 HR: 75.1 (95% CI: 61.7, 88.4) vs. 16.4 (95% CI: 12.8, 20.0)). This changed to either no difference or higher risk for the VHA-use group as time since separation increased. For unintentional overdose, there was no difference in HRs between groups in 0-2 years post-separation; risk was higher for the VHA-use group for the remainder of follow-up (e.g., no VHA use, years 3-5 HR: 8.7 (95% CI: 5.1, 12.3) vs. 28.9 (95% CI: 25.2, 32.5) for VHA use group). CONCLUSIONS: The risk for unintentional MVC and overdose death since separation differed based on use of VHA services among post-9/11 Veterans. Intervention strategies that address differences in time since separation and whether Veterans use the VHA or not may be required.