Ultra-short screening instruments for major depressive episode and generalized anxiety disorder in epilepsy: The NDDIE-2 and the GAD-SI

Highlights

• An ultra-short version of the NDDI-E (2 items) is effective to screen MDE in PWE.

• An ultra-short version of the GAD-7 (1 item) is effective to screen GAD in PWE.

• Ultra-short versions could be advantageous for patients with cognitive impairment.

• Further studies are needed in larger populations and in multiple care settings.

Abstract

Background

Systematic screening is recommended for major depressive episode (MDE) with the Neurological Disorders Depression Inventory for Epilepsy NDDI-E, 6 items and generalized anxiety disorder (GAD) with the GAD 7 items in patients with epilepsy (PWE). Shorter versions of the NDDI-E and the GAD-7 could facilitate increased screening by busy clinicians and be more accessible to patients with mild cognitive and/or language impairments.

Methods

The effectiveness of ultra-short versions of the NDDI-E (2 items) and the GAD-7 (the GAD-2, 2 items, and the GAD-SI with a single item) in comparison with the original versions were statistically tested using ROC analysis.

Results

ROC analysis of the NDDIE-2 showed an AUC of 0.926 (p<0.001), a sensitivity of 81.82% and a specificity of 89.16%, without significant difference with the NDDI-E (z=1.582, p=0.11). ROC analysis of the GAD-SI showed an AUC of 0.872 (p<0.001), a sensitivity of 83.67% and a specificity of 82.29%, without significant difference with the GAD-7 (z=1.281, p=0.2). The GAD-2 showed poorer psychometric properties.

Limitations

The limitation is the use of data from previously reported subjects in a single language version, the NDDIE-2 that lacks detection of dysphoric symptoms in comparison with the NDDIE-6 and the GAD-SI that exhibited a more than 10% lower sensitivity than the GAD-7.

Conclusions

This study highlights the potential utility of the NDDIE-2 and the GAD-SI as ultra-short screening tools for MDE and GAD respectively in PWE. Further studies in a larger population, including multi-lingual versions, could be a valuable next step. However, the brevity and simplicity of this tool could be an advantage in PWE who present cognitive difficulties, especially attentional or language deficits.

Introduction

Major depressive episode (MDE) and generalized anxiety disorder (GAD) are the two most prevalent psychiatric disorders in persons with epilepsy (Kwon and Park, 2014). Given effects on epilepsy prognosis, quality of life and mortality, systematic screening is recommended for MDE and GAD in this population. To this aim 2 short self-rated questionnaires, validated in multiple languages, are promoted by the International League Against Epilepsy (ILAE) Commission on Neuropsychiatry (Gilliam et al., 2006, Micoulaud-Franchi et al., 2015). The Neurological Disorders Depression Inventory for Epilepsy (NDDI-E) is a 6-item self-rated questionnaire on a balanced four point Likert scale ranging from 1 to 4 (Gilliam et al., 2006, Micoulaud-Franchi et al., 2015). For clarity, we will refer to the original NDDI-E as the “NDDIE-6″. The Generalized Anxiety Disorder 7 (GAD-7) is a 7-item self-rated questionnaire on a balanced four point Likert scale ranging from 0 to 3 (Micoulaud-Franchi et al., 2016, Ruiz et al., 2011). Within the spectrum of screening instruments, the NDDIE-6 and the GAD-7 are considered “short screening tools” (i.e. 5–14 items, taking between 2 and 5 min to complete) as opposed to “standard screening tools” (i.e. 15 or more items, >5 min to complete) (Mitchell and Coyne, 2007).

Though short, these tools may still seem too lengthy and under-utilized in busy clinical practice (Kroenke et al., 2003, Kroenke et al., 2007). Epilepsy patients may have cognitive deficits making even relatively short questionnaires challenging to complete. Diverse ethnic case-mix in epilepsy care giving rise to variable language abilities has also been highlighted (Rampling et al., 2012).

So-called “ultra-short screening instruments” (1–4 items, <2 min to complete) have been validated in the general population (Kroenke et al., 2003, Kroenke et al., 2007) and some specific clinical conditions (Seo and Park, 2015a, Seo and Park, 2015b). The best known is the PHQ-2 (Kroenke et al., 2003, Mitchell and Coyne, 2007, Whooley, 2016), a self-reported screening questionnaire for depression based on only two questions, promoted in national guidelines for widespread use by both the National Institute of Health in the United States and the National Institute for Clinical Excellence in the United Kingdom. A community-based study of persons with epilepsy comparing the NDDIE-6 and the PHQ-2 found good positive and negative predictive values for both scales, and good patient acceptability (Margrove et al., 2011). However, a large tertiary hospital study found poor sensitivity with the PHQ-2, suggesting that it should not be recommended for routine screening use in persons with epilepsy (Fiest et al., 2014). A 2 item screening tool (GAD-2) and a single item screening tool (GAD-SI) have also been validated for screening for GAD (Donker et al., 2011, Kroenke et al., 2007), but not to date studied in persons with epilepsy.

The present study aims to analyze the potential effectiveness of ultra-short versions of the NDDIE-6 and the GAD-7 in screening for MDE and GAD respectively, in a sample of persons with epilepsy in a tertiary specialist epilepsy service.