Abstracts

Rationale: Previous studies have demonstrated depression in about a third of community-based epilepsy patients. A prior study compared results of the CES-D and NDDI-E in patients from tertiary epilepsy centers, however little is known about this comparison in community based epilepsy patients. The purpose of this study is to estimate the prevalence of comorbid depression in epilepsy and assess the extent of agreement by two validated depression instruments (CES-D and NDDI-E) in a larger and broader epilepsy sample. Methods: A cross-sectional survey was conducted of adult patients with epilepsy. Patients were identified from a large US health plan, and were required to have two or more diagnoses for epilepsy (ICD-9-CM 345.xx) or non-febrile convulsions (ICD-9-CM 780.3 or 780.39) on two separate dates of service as well as at least two pharmacy claims for AED medications on different dates of service. Two validated measures were used to assess depression: the CES-D and the NDDI-E. Data were collected on demographic and clinical characteristics, quality of life, cognitive function, seizure severity, medication adherence, and adverse events. All survey variables were analyzed descriptively. Results: Data were analyzed from 465 adult patients with epilepsy. Patients were 45% male, mean age of 44.77 years (SD 13.41), and 93% white. Based on the CES-D, 36% were categorized as depressed with a mean score of 25.53 (SD 8.61). The NDDI-E categorized 19% as depressed with a mean score of 18.31 (SD 1.96). After comparison of prevalence estimates of depression between the CES-D and NDDI-E, 63% of patients were categorized as not depressed by both; 18% categorized as depressed by CES-D but not depressed by NDDI-E; 1% depressed by NDDI-E but not depressed by CES-D; and 17% depressed according to both. Based on the CES-D, there were significant differences in gender and employment such that males were less likely to be depressed (p=0.009) and those not employed were more likely to be depressed (p<0.001). Employment status (p=0.002) was also significant based on the NDDI-E. Conclusions: These findings suggest a high prevalence of depression in epilepsy, as validated by both the CES-D and NDDI-E. The low degree of overlap between the two scales indicates that the NDDI-E may be identifying a subset of patients among those considered depressed by the CES-D. This should be taken into account in future studies and choice of instrument to identify depression in patients with epilepsy.