Objectives: Although Veterans Affairs (VA) patients with posttraumatic stress disorder (PTSD) are prescribed benzodiazepines and opioids in addition to recommended pharmacotherapies, little is known about the safety of these medications. This study compared the 2-year incidence of adverse events among VA patients with PTSD exposed to combinations of selective serotonin reuptake inhibitors (SSRIs) or serotonin/norepinephrine reuptake inhibitors (SNRIs), benzodiazepines, and opioids.
Methods: This retrospective cohort study used VA administrative data from 2004 to 2010 to identify and follow 5236 VA patients with PTSD with new episodes of (1) SSRIs/SNRIs only; (2) concurrent SSRIs/SNRIs and benzodiazepines; and (3) concurrent SSRIs/SNRIs, benzodiazepines, and opioids. Outcome measures were the 2-year incidence and adjusted hazard ratios (AHR) of mental health and medicine/surgery hospitalizations, emergency department visits, harmful events (eg, injuries and death), and any adverse event after adjustment for demographics, clinical covariates, and adverse event history.
Results: Compared with SSRIs/SNRIs only, the adjusted risk of mental health hospitalizations (AHR: 1.87; 95% confidence interval [CI]: 1.37-2.53) was greater among patients prescribed SSRIs/SNRIs and benzodiazepines concurrently. The AHR of mental health hospitalizations (AHR: 2.00; 95% CI: 1.35-2.98), medicine/surgery hospitalizations (AHR: 4.86; 95% CI: 3.30-7.14), emergency department visits (AHR: 2.01; 95% CI: 1.53-2.65), any harmful event (2.92; 95% CI: 2.21-3.84), and any adverse event (AHR: 2.65; 95% CI: 2.18-3.23) were all significantly greater among patients prescribed SSRIs/SNRIs, benzodiazepines, and opioids than among those prescribed SSRIs/SNRIs only.
Conclusions: Concurrently prescribing SSRIs/SNRIs, benzodiazepines, and opioids among patients with PTSD is associated with adverse events. Although efforts are warranted to monitor patients who are prescribed combinations of these medications to prevent adverse events, these results should be interpreted with caution until they are replicated.