Rapid detection of major depression in epilepsy: a multicentre study

Abstract

Background Depression is a common comorbid disorder in epilepsy but is not routinely assessed in neurology clinics. We aimed to create a rapid yet accurate screening instrument for major depression in people with epilepsy.Methods We developed a set of 46 items to identify symptoms of depression that do not overlap with common comorbid cognitive deficits or adverse effects of antiepileptic drugs. This preliminary instrument and several reliable and valid instruments for diagnosis of depression on the basis of criteria from the Diagnostic and Statistical Manual IV, depression symptom severity, health status, and toxic effects of medication were applied to 205 adult outpatients with epilepsy. We used discriminant function analysis to identify the most efficient set of items for classification of major depression, which we termed the neurological disorders depression inventory for epilepsy (NDDI-E). Baseline data for 229 demographically similar patients enrolled in two other clinical studies were used for verification of the original observations.Findings The discriminant function model for the NDDI-E included six items. Internal consistency reliability of the NDDI-E was 0.85 and test-retest reliability was 0.78. An NDDI-E score of more than 15 had a specificity of 90%, sensitivity of 81%, and positive predictive value of 0.62 for a diagnosis of major depression. Logistic regression showed that the model of association of major depression and the NDDI-E was not affected by adverse effects of antiepileptic medication, whereas models for depression and generic screening instruments were. The severity of depression symptoms and toxic effects of drugs independently correlated with subjective health status, explaining 72% of variance. Results from a separate verification sample also showed optimum sensitivity, specificity, and predictive power at a cut score of more than 15.Interpretation Major depression in people with epilepsy can be identified by a brief set of symptoms that can be differentiated from common adverse effects of antiepileptic drugs. The NDDI-E could enable rapid detection and improve management of depression in epilepsy in accordance with internationally recognised guidelines.