Using the Minnesota Multiphasic Inventory 2, EEGs, and clinical data to predict nonepileptic events

Minnesota Multiphasic Inventory 2 (MMPI-2) scale 3, duration of illness, and routine EEGs have been used to predict nonepileptic events (NEEs) with a high degree of accuracy in patients referred for video/EEG (vEEG) monitoring. This study tested the Storzbach logistic regression equation in our patients with definitive epileptic seizures (n = 57) or NEEs without evidence of epileptiform activity (n = 51) during vEEG monitoring, yielding an overall classification accuracy of 81%, sensitivity of 80%, and specificity of 81%. This study also replicated previous findings of significant group differences in duration (years) of spells, number of elevations on the MMPI-2, MMPI-2 elevations on scales 1, 2, 3, and 8, and incidence of the conversion valley on the MMPI-2. Our findings indicated that combined use of the MMPI-2 and clinical variables was most predictive of patients with NEEs.