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

Area of Science:

• Neurology
• Psychiatry
• Clinical Psychology

Background:

• Predicting nonepileptic events (NEEs) in patients undergoing video/EEG (vEEG) monitoring is crucial for accurate diagnosis and treatment.
• Previous studies identified Minnesota Multiphasic Inventory 2 (MMPI-2) scale 3, illness duration, and routine EEGs as predictors of NEEs.

Purpose of the Study:

• To validate the Storzbach logistic regression equation for predicting NEEs in a patient cohort undergoing vEEG monitoring.
• To assess the combined predictive value of the MMPI-2 and clinical variables for identifying NEEs.

Main Methods:

• A logistic regression equation was tested on patients with confirmed epileptic seizures (n=57) and NEEs (n=51) during vEEG monitoring.
• Statistical analysis compared groups based on MMPI-2 results (scale elevations, conversion valley) and clinical variables (illness duration).

Main Results:

• The Storzbach equation achieved an overall classification accuracy of 81%, with 80% sensitivity and 81% specificity for NEEs.
• Significant group differences were found in illness duration, MMPI-2 elevations (scales 1, 2, 3, 8), and the incidence of the conversion valley.

Conclusions:

• The combined use of the MMPI-2 and clinical variables demonstrated the highest predictive accuracy for identifying patients with NEEs.
• These findings support the utility of integrating psychological assessment with clinical data for diagnosing NEEs.