Brief report
Measuring depression outcome with a brief self-report instrument: sensitivity to change of the Patient Health Questionnaire (PHQ-9)

https://doi.org/10.1016/S0165-0327(03)00198-8Get rights and content

Abstract

Background: The nine-item depression module from the Patient Health Questionnaire (PHQ-9) is well validated and widely used as a brief diagnostic and severity measure, but its validity as an outcome measure for depression has not yet been established. Therefore, we investigated the sensitivity to change of the PHQ-9 in three groups of patients whose depression status either improved, remained unchanged, or deteriorated over time. Methods: From three cohorts of medical outpatients, with an equal distribution of major depressive disorder, other depressive disorders, or no depressive disorder, 167 patients (82.7%) were followed up after a mean of 12.3±3.0 months. The PHQ-9 and the Structured Clinical Interview for DSM-IV (SCID) were completed at both baseline and follow-up. Depression diagnoses from the SCID were used as the criterion standard to divide patients into subgroups with (a) improved depression status, (b) unchanged depression status, and (c) deteriorated depression status. Results: Effect sizes (ES) of PHQ-9 change scores were ES=−1.33 for the improved depression status subgroup (n=52), ES=−0.21 for the unchanged status subgroup (n=91), and ES=0.47 for the deteriorated status subgroup (n=24). PHQ-9 change scores differed significantly between the three depression outcome groups. Limitations: The PHQ-9 and the SCID were completed in person at baseline, whereas they were completed in a telephone interview at follow-up. Conclusions: This study demonstrates the ability of the PHQ-9 to detect depression outcome and changes over time. Data from treatment trials will help further establish the sensitivity to change of the PHQ-9 in comparison to other depression severity measures.

Introduction

Many depression screening instruments have been developed during the past 40 years. However, the “Patient Health Questionnaire (PHQ)” (Spitzer et al., 1999) was the first self-report questionnaire designed for use in primary care that actually diagnoses specific disorders using criteria from the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) (American Psychiatric Association, 2000). Its nine-item depression module, the PHQ-9, is increasingly being used as a brief diagnostic and severity measure in research and clinical practice (Kroenke et al., 2001). The superior criterion validity of the PHQ-9 compared to two other established depression screening questionnaires has recently been demonstrated with respect to the diagnosis of ‘major depressive disorder’ made by a standard interview in assessing psychiatric disorders (Löwe et al., in press, a). Similarly, superior operating characteristics were found for the panic module of the PHQ (Löwe et al., in press, b).

Apart from criterion validity, the most important characteristic of an outcome measure is to appropriately reflect change over time, for instance, the responsiveness to a specific treatment. However, sensitivity to change of the PHQ-9 has not yet been evaluated. If it does prove sensitive to change, the PHQ-9 would be a practical option for assessing depression outcomes in research and clinical practice.

To examine the sensitivity to change of the PHQ-9, we prospectively followed cohorts of medical outpatients with major depressive disorder, other depressive disorders, or no depressive disorder. The PHQ-9 and a standard diagnostic interview were completed at baseline and follow-up. Using the diagnostic interview as the external criterion standard, the patients were assigned to subgroups of improved, unchanged, or deteriorated depression status. With regard to sensitivity to change, we hypothesised that PHQ-9 change scores would differ significantly between these three patient subgroups.

Section snippets

Subjects

From the baseline sample of the German PHQ validation study1 (Löwe et al., in press, a), the following patients were contacted for the follow-up assessment after a mean of 12.3±3.0 months: (i) any patient with ‘major depressive disorder’ diagnosed using the

Results

At follow-up, ‘major depressive disorder’, according to SCID, was present in 40 patients (24.0%), ‘other depressive disorders’ in 43 patients (25.7%), and ‘no depressive disorder’ in 84 patients (50.3%). As a result, a total of 52 patients (31.1%) improved in SCID depression diagnostic status from their baseline to their follow-up assessment, 91 patients (54.5%) remained stable, and 24 patients (14.4%) deteriorated. The corresponding PHQ-9 scores and sensitivity to change indices of the patient

Discussion

This study was performed with the primary objective of evaluating the sensitivity to change of a brief depression diagnostic and severity instrument. We found that changes in the PHQ-9 score correspond with changes in depression diagnostic status over time, providing preliminary evidence that the PHQ-9 can be used for longitudinal as well as for cross-sectional studies. The patient groups were well matched and had excellent follow-up rates.

Our study has several limitations. First, the PHQ-9 and

Acknowledgments

This study was supported by unrestricted research grants from Pfizer, Germany and from the medical faculty of the University of Heidelberg, Germany (project 121/2000). There are no conflicts of interest.

We would like to thank our patients for their participation at the baseline and the 1-year follow-up assessment. Moreover, we would also like to express our thanks to Dipl.-Psych. Beate Wild and to Dipl.-Psych. Dieter Schellberg for their helpful suggestions regarding data analyses and to cand.

References (19)

There are more references available in the full text version of this article.

Cited by (1187)

View all citing articles on Scopus
View full text