Depression in people with epilepsy: reliability and validity of the Czech version of Beck Depression Inventory (BDI-II) and Beck Depression Fast Screen (BDI-FS) for major depressive episode

Abstract

Objective Depression affects up to 55% of people with epilepsy (PWE) yet remains systematically underdiagnosed. This study aimed to evaluate the psychometric properties of the Czech versions of the Beck Depression Inventory–II (BDI-II) and Beck Depression Inventory–Fast Screen (BDI-FS) for identifying major depressive episode (MDE) in PWE. Methods A cross-sectional study was conducted with n = 278 adult PWE, assessed at the tertiary epilepsy center. MDE diagnosis was established using the Mini International Neuropsychiatric Interview (M.I.N.I. 6.0.0) as the diagnostic gold standard. Psychometric evaluation included internal consistency, test–retest reliability, exploratory factor analysis, diagnostic accuracy via ROC analysis and concurrent validity against the The Neurological Disorders Depression Inventory for Epilepsy (NDDI-E), The Functional Activities Questionnaire (FAQ), and QOLIE-89. Results 25% of the sample (n = 70) met criteria for MDE. Both instruments demonstrated excellent reliability (BDI-II: α = 0.93; BDI-FS: α = 0.81) and strong one-month test–retest stability (ICC = 0.84 and 0.79, respectively). Factor analysis replicated a two-factor structure comprising Somatic–Affective and Cognitive dimensions, accounting for 44% of total variance. Both scales correlated strongly with the NDDI-E (ρ ≈ 0.74) and showed expected negative strong correlations with functional capacity and quality of life. ROC analysis identified optimal cutoffs of BDI-II > 14 (AUC = 0.984, sensitivity 95.7%, specificity 95.1%) and BDI-FS > 3 (AUC = 0.925, sensitivity 84.3%, specificity 85.6%). Conclusion The Czech BDI-II version and its brief alternative BDI-FS are reliable and valid instruments for screening MDE in epilepsy. Implementation of these tools in epilepsy clinics may substantially improve depression detection and patient outcomes.