Abstracts

Rationale: Objective: To evaluate mortality in elderly veterans newly diagnosed epilepsy versus those with an existing diagnosis of epilepsy and compare them across comorbid conditions. Background: It is well established that individuals with epilepsy have elevated mortality. Factors associated with increased mortality include age, geographic location, epilepsy etiology, seizure type & severity, medication burden, and comorbid conditions. However, there is a paucity of population based studies in elderly patients, who represent the fastest growing segment of general population and have an increased incidence of new onset epilepsy. In a previous study of elderly veteran patients with a new diagnosis of epilepsy we found an extremely high absolute mortality of 12% at 1 year, and 32.9% within 3 years. In the current study we looked at mortality in elderly veteran patients with either a new diagnosis of epilepsy or a pre-existing (chronic) epilepsy diagnosis and the association with comorbid diagnoses. . Methods: We used VA administrative data to identify veterans 66+ years who received VA care in FY05 and FY06. Individuals were classified as epilepsy cohort or the non-epilepsy cohort based on a previously validated algorithm. Our primary measure was mortality within the cohort groups, which was compared to a no epilepsy cohort using the VA minivital status file. In addition,we identified covariates of that included patient demographics (age, gender, race); disease burden; and comorbid conditions. Results: For FY 2006, we identified 35,611 veterans with chronic epilepsy and 1,412 veterans with a new diagnosis of epilepsy. In the chronic epilepsy group there were 2,468 deaths (6.9%) and in the new epilepsy group there were 131 deaths (9.3%) which is almost double the one year mortality seen in the no epilepsy cohort (4.1%). The majority of the mortality was associated comorbid disease with the highest associations with cancer, various cardiac conditions and dementia. Conclusions: As has been previously demonstrated, the presence of epilepsy is associated with elevated mortality amongst elderly veterans and is greater for patient's with a new diagnosis of epilepsy as opposed to a pre-existing diagnosis. Although mortality is elevated for both groups. However, when compared with age matched veterans , who do not have epilepsy, the majority of the mortality variance appears to be accounted for by the presence of comorbid medical conditions.