Background: The Improving Access to Psychological Therapies (IAPT) program uses the Patient Health Questionnaire (PHQ-9), the Generalized Anxiety Disorder Scale (GAD-7), and the Work and Social Adjustment Scale (WSAS) as part of their unique outcome monitoring system. To reduce patient burden, this study explored whether abbreviated versions of these questionnaires can be used to derive relevant outcome statistics with minimal loss of information.
Methods: Using two samples (training; n = 1530, validation; n = 766), we examined whether existing short-forms, PHQ-4 and GAD-R3, would provide enough information to calculate relevant outcomes with near perfect agreement with the outcomes based on the original scales. We also examined 1) whether additional items would further improve the agreement between the abbreviated and original scales, and 2) alternative short-forms based on the sample-derived item information curves. The latter was also used to derive an abbreviated version of WSAS.
Results: The abbreviated version derived from the item information curves provided the closest match with the original scales. A 5-item version of PHQ, a 4-item version of GAD, and a 3-item version of WSAS were correlated 0.95 with their original counterpart. Agreement as expressed by Cohen's kappa also suggested near perfect agreement for the outcomes (reliable) recovery rate and reliable improvement rate (>0.80). The outcome point estimates also matched very well (<2 % difference). Results were replicated in the validation sample.
Conclusion: The derived abbreviated versions can be used for the purpose of routine outcome monitoring with minimal loss of information and reduce patient burden with nearly 50 %.
Keywords: Anxiety; Depression; Functioning; Measurement; Outcome monitoring; Patient burden.
Copyright © 2025 The Authors. Published by Elsevier B.V. All rights reserved.