Screening for depression in epilepsy clinics. A comparison of conventional and visual-analog methods

Purpose: Depression is an important but underdiagnosed complication of epilepsy. This study compares potentially suitable screening tools head-to-head.

Methods: We enrolled 266 attendees with a confirmed diagnosis of epilepsy at a specialized neurologic epilepsy service in London and compared verbal self-report and visual analog (VAS) screening methods for depression. These included two generic depression scales (Hospital Anxiety and Depression Scale [HADS], Beck Depression Inventory II [BDI-II]), one epilepsy specific scale (Neurological Disorders Depression Inventory for Epilepsy [NDDI-E]) and one new visual-analog scale (Emotional Thermometers [ET]). We used Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria for major depression and International Classification of Diseases, Tenth Revision (ICD-10) criteria for depressive episode as the reference standard.

Key findings: Against ICD-10-defined depression the most accurate scales by receiver operating characteristic (ROC) curve area were HADS Total (HADS-T, 0.924), BDI-II (0.898) and NDDI-E (0.897). New visual-analog methods had similar accuracy measured either in combination or individually. Although no test performed well in a case-finding role, several performed well as a rule-out initial step, owing to high negative predictive value and specificity. In this role, the optimal performing conventional tools were the HADS depression subsscale (HADS-D) and the NDDI-E and the optimal single VAS were the depression thermometer (DepT) and the distress thermometer (DT). Against DSM-IV- defined major depression, results were similar with optimal performance by the HADS-T, BDI-II, and NDDI-E, but here the anxiety thermometer (AnxT) as well as DepT and DT also offered good performance. Given that no test performed well in a case-finding role, we suggest that these tests are used as an initial first step to rule out patients who are unlikely to have depression.

Significance: We suggest that the six-item NDDI-E or seven-item HADS-D should be considered if a conventional scale is preferred and that the revised ET4 be considered if a visual-analog method is required. Follow-up examination and intervention, where indicated, are necessary in all those who screen positive on any measure as these are not intended as diagnostic tools.