Abstracts

Rationale: Health Related Quality of Life (HRQoL) is a multidimensional construct known to be negatively affected by both epileptic seizures (ES) and psychogenic non-epileptic seizures (PNES). U.S. veterans are a distinct patient group with regard to gender, age, and background.  We studied HRQoL in a well characterized group of veterans with ES or PNES and asked three questions: (1) Is there a difference in HRQoL in veterans with ES vs. PNES? (2) What factors influence HRQoL in each group? (3) What factors influence the difference in HRQoL between groups? Methods: The COVE (Characteristics of Veterans with Epilepsy) study prospectively evaluated 333 veterans entering the epilepsy monitoring units of the Portland Oregon, San Francisco California, and Madison Wisconsin VA Epilepsy Centers of Excellence. All underwent continuous video-EEG monitoring. Final seizure diagnoses followed established criteria. For this study we included patients diagnosed with ES (n=64) or PNES (n=73). HRQoL was measured with the Quality of Life in Epilepsy-31 (QOLIE). Psychiatric diagnoses were obtained using the Structured Clinical Interview for DSM IV. Severity of current PTSD symptoms was measured with the PTSD checklist (PCL), and current depression symptoms by the Beck Depression Inventory II (BDI-II). The MMPI-2RF RC1 scale (RC1) measured somatic complaints.To assess the influence of several factors on HRQoL we performed hierarchical multiple regression. Candidate variables were included in the analysis if correlations with the QOLIE total score were p<0.1 for the ES or PNES groups. We included categorical measures meeting a p<0.1 threshold on a 2-group Wilcoxon test. Measures were entered in the following order: demographic, social, seizure-related, psychological. Adjusted R²values were calculated separately for ES and PNES groups. Factors contributing to the between-group difference in QOLIE scores were evaluated using a change-in-estimate strategy (estimated standardized coefficients).   Results: QOLIE scores were higher in the ES vs. PNES group (50.9 vs. 39.0; p<0.001). There were significant differences (p<0.01) in 5 of 7 subscales. Regression analysis revealed minor contributions of demographic, social, and seizure factors to the QOLIE total score (cumulative adjusted R² of 0.18 and 0.07 for ES and PNES groups). Inclusion of psychological scores (BDI-II, PCL, RC1, number of axis 1 diagnoses) increased the adjusted R² to 0.66 and 0.65. Regression for PCL scores reduced the estimated standardized coefficient for seizure type (ES vs. PNES) to zero, and regression for either RC1 or BDI-II scores reduced the estimated coefficient by >70%. After accounting for psychological variables seizure type did not have a significant effect on QOLIE scores. Conclusions: QOLIE-31 scores were higher in veterans with ES vs. PNES. In both groups HRQoL was largely driven by psychological factors; demographic, military, social, and seizure related factors had minimal influence. The difference in QOLIE scores between ES and PNES groups was due to psychological factors, current PTSD symptoms having the strongest effect. Funding: Supported by grant 5101cx000721-04 from the US Department of Veterans Affairs.