A systematic review and meta-analysis of diagnostic test accuracy studies of self-report screening instruments for common mental disorders in Arabic-speaking adults.

Anne M De Graaff,Ralph De Vries,Irene Sferra,Marit Sijbrandij,Mariska Leeflang,Pim Cuijpers,Jana R Uppendahl

doi:10.1017/gmh.2021.39

Abstract

Self-report screening instruments are frequently used as scalable methods to detect common mental disorders (CMDs), but their validity across cultural and linguistic groups is unclear. We summarized the diagnostic accuracy of brief questionnaires on symptoms of depression, anxiety and posttraumatic stress disorder (PTSD) among Arabic-speaking adults. Five databases were searched from inception to 22 January 2021 (PROSPERO: CRD42018070645). Studies were included when diagnostic accuracy of brief (maximally 25 items) psychological questionnaires was assessed in Arabic-speaking populations and the reference standard was a clinical interview. Data on sensitivity/specificity, area under the curve, and data to generate 2 × 2 tables at various thresholds were extracted. Meta-analysis was performed using the diagmeta package in R. Quality of studies was assessed with QUADAS-2. Thirty-two studies (N participants = 4042) reporting on 17 questionnaires with 5-25 items targeting depression/anxiety (n = 14), general distress (n = 2), and PTSD (n = 1) were included. Seventeen studies (53%) scored high risk on at least two QUADAS-2 domains. The meta-analysis identified an optimal threshold of 11 (sensitivity 76.9%, specificity 85.1%) for the Edinburgh Postnatal Depression Scale (EPDS) (n studies = 7, n participants = 711), 7 (sensitivity 81.9%, specificity 87.6%) for the Hospital Anxiety and Depression Scale (HADS) anxiety subscale and 6 (sensitivity 73.0%, specificity 88.6%) for the depression subscale (n studies = 4, n participants = 492), and 8 (sensitivity 86.0%, specificity 83.9%) for the Self-Reporting Questionnaire (SRQ-20) (n studies = 4, n participants = 459). We present optimal thresholds to screen for perinatal depression with the EPDS, anxiety/depression with the HADS, and CMDs with the SRQ-20. More research on Arabic-language questionnaires, especially those targeting PTSD, is needed.

Highlights

Common mental disorders (CMDs) such as depression, anxiety, and posttraumatic stress disorder (PTSD) affect millions of people globally
There is a large variety of brief, self-report screening instruments for symptoms of common mental disorders (CMDs), such as the Hopkins Symptoms Checklist (HSCL), the Hospital Anxiety and Depression Scale (HADS), and the PTSD Checklist (PCL)
The sensitivity in seven studies evaluating the Edinburgh Postnatal Depression Scale (EPDS) ranged from 73% to 92%; its specificity ranged from 48% to 96%

Summary

Introduction

Common mental disorders (CMDs) such as depression, anxiety, and posttraumatic stress disorder (PTSD) affect millions of people globally. There is a large variety of brief, self-report screening instruments for symptoms of CMDs, such as the Hopkins Symptoms Checklist (HSCL), the Hospital Anxiety and Depression Scale (HADS), and the PTSD Checklist (PCL). The ease of administration of most self-report measures makes them attractive for use in research (Kagee et al, 2013) These instruments are usually developed and evaluated in specific (Western, Anglo-Saxon) settings (Saxena et al, 2006; Ali et al, 2016), while psychometric properties may vary across settings, cultures, and languages. Self-report screening instruments are frequently used as scalable methods to detect common mental disorders (CMDs), but their validity across cultural and linguistic groups is unclear. More research on Arabiclanguage questionnaires, especially those targeting PTSD, is needed

Methods

Results

Conclusion