Diagnosing ICD-10 Depressive Episodes: Superior Criterion Validity of the Patient Health Questionnaire

Vol. 73, No. 6, 2004

Free Abstract Article (Fulltext) Article (PDF 87 KB)

Clinical Note

Diagnosing ICD-10 Depressive Episodes: Superior Criterion Validity of the Patient Health Questionnaire
Bernd Löwe^a,b, Kerstin Gräfe^b, Stephan Zipfel^b, Steffen Witte^c, Bernd Loerch^d, Wolfgang Herzog^b

^aRegenstrief Institute, Indiana University School of Medicine, Indianapolis, Ind., USA;
^bDepartment of General Internal and Psychosomatic Medicine, University of Heidelberg, Medical Hospital,
^cDepartment of Medical Biometry, University of Heidelberg, Medical Faculty, Heidelberg, and
^dDepartment of Psychiatry, University of Mainz, Mainz, Germany

Address of Corresponding Author

Psychother Psychosom 2004;73:386-390 (DOI: 10.1159/000080393)

Key Words

Depressive disorder
Diagnosis
Questionnaires
Sensitivity and specificity
Reproducibility of results
Primary health care
Receiver operating characteristic curve
Patient Health Questionnaire

Abstract

Background: Diagnosing and monitoring depression in primary care remains an issue of significant public health concern. Clinicians and researchers need to know if any one screening instrument is superior to the others in diagnosing ICD-10 depressive episodes. This study aimed to examine the criterion validity for diagnosing ICD-10 depressive episodes of the Patient Health Questionnaire (PHQ) in comparison with 2 well-established instruments, the Hospital Anxiety and Depression Scale (HADS), and the WHO Well-Being Index 5 (WBI-5). Methods: Five hundred and one medical outpatients completed the questionnaires and had a clinical interview. The presence of a depressive episode was determined with the International Diagnostic Checklists (IDCL) for ICD-10 as the criterion standard. Coefficient kappa ( kappa ), sensitivities and specificities were calculated and a statistical comparison of the areas under the receiver operating characteristic curves was performed. Results: Diagnostic agreement between the questionnaires and the IDCL was moderate ( kappa = 0.34-0.56), with the highest values for the PHQ. While all 3 questionnaires had reasonable sensitivity and specificity, the operating characteristics for the PHQ were significantly superior to both the HADS and the WBI-5 (p = 0.02). Conclusions: Any of the 3 screening instruments can be recommended for clinical use. However, this is the first comparative study to demonstrate the diagnostic advantage of a particular depression-screening instrument using the ICD-10 diagnostic criteria. The superior criterion validity of the PHQ is likely attributable to its closer representation of the current concept of depressive disorders.

Author Contacts

Bernd Löwe, MD, PhD
Department of General Internal and Psychosomatic Medicine
Heidelberg University Medical Center, Im Neuenheimer Feld 410
DE-69120 Heidelberg (Germany)
Tel. +49 6221 568999, Fax +49 6221 565749, E-Mail bernd.loewe@med.uni-heidelberg.de

Article Information

Number of Print Pages : 5
Number of Figures : 1, Number of Tables : 2, Number of References : 18

Medline Abstract (ID 15479995)

Download Citation

For non-native English speakers and international authors who would like assistance with their writing before submission, we suggest American Journal Experts for their scientific editing service.