Using the Edinburgh postnatal depression scale to screen for anxiety disorders: Conceptual and methodological considerations

Background: Perinatal anxiety symptoms and disorders are prevalent and disabling but have not to date been a focus for specific clinical and public health attention. The EPDS is widely used to detect probable depression, and many studies have also found that three items from this scale load on an anxiety factor, in both the antenatal and postnatal periods. In addition, studies have found clinically significant correlations between the EPDS and various anxiety-specific measures in the perinatal period. The aim of this paper is to examine studies which address the capacity of the EPDS to detect anxiety disorders, to assess whether the EPDS performs differently in women with depressive or anxiety disorders and to consider the implications for future research and clinical practice. Methods: The English-language perinatal mental health literature was searched. Six studies with data pertaining to the capacity of the EPDS to detect perinatal anxiety disorders in women were identified. These studies provide information on i) comparison of total EPDS score by diagnoses of anxiety and depression and ii) comparison of the anxiety subscale score (EPDS-3A) by diagnoses of anxiety and depression. Results: There is evidence from both sets of information that the EPDS is useful for screening for anxiety in women and emerging evidence that Total EPDS and EPDS-3A can distinguish depression from anxiety reliably. Limitations: The findings are based on a small number of studies, conducted in a variety of clinical and community settings in different languages and countries, and with variable sample sizes, some of which lack power to ensure reliable conclusions. Conclusions: The EPDS appears to detect perinatal anxiety disorders, but further research is required to establish the clinical and public health value of the EPDS for this purpose, and whether it has more robust psychometric properties or is more feasible and acceptable than existing anxiety-specific measures. Crown Copyright (C) 2012 Published by Elsevier B.V. All rights reserved.