Abstracts

Rationale: Epilepsy, advancing age and being a veteran are all factors associated with an increase risk of suicide in the US. Early in 2008, the FDA issued an alert indicating that antiepileptic drug (AED) treatment is associated with an increase risk of suicidal thoughts and behavior. Our objective was to identify risk factors associated with suicide in older veterans taking AEDs. Methods: National VA databases were used to identify a geriatric cohort aged ≥66 years in FY00-04 who received VA care for at least one year prior to being treated with an AED for any condition. We identified co-morbid conditions before and after treatment and studied AED treatment persistence. Age matched controls were compared to veterans who committed suicide after initiation of an AED. Demographic and clinical conditions were identified with validated ICD-9-CM codes. Results: 64 suicide cases were compared to 768 controls; the majority (98%) were men and 66-84 yo (97%). The primary predictor of suicide (matched on year of first AED treatment, epilepsy status and prior suicide attempt) was the diagnosis of depression (OR 4.42, 95% CI 2.30-8.49) before treatment with an AED. Of the AEDs, valproic acid was more common (14% vs 5%) in the suicide group , but based on multivariable analysis there were no significant differences between the groups and the different AEDs. Epilepsy, dementia, chronic pain and numerous co-morbid medical conditions were not more common in the suicide group Conclusions: Our study found that in older veterans the most important predictive factor of suicide was a psychiatric diagnosis of depression prior to AED treatment. These findings suggest that screening and intervention for depression and suicidal ideation are paramount in the prevention of suicide in geriatric veterans treated with AEDs.