Skip to main content

Personal recovery as outcome measure in mental health research: A meta-analysis on the sensitivity to change properties of the Recovery Assessment Scale.

Journal of affective disorders

Authors: T A D Pelgrim, K Kleine Schaars, R Moldovan, M Deen, R van Westrhenen, Kristian Kleine Schaars, Sanah Bedi, Allan H Young, Mario Juruena, Daniel Silman, Urs Heilbronner, Monika Budde, Magnus Ingelman-Sundberg, Marin Jukic, Martien J H Kas, Raj Jagesar, Ramona Moldovan, Markus M Nöthen, Per Hoffmann, Carina M Mathey, Alexandra Philipsen, Laura L Kilarski, Jonathan Laatsch, Noam Shomron, Erik Van der Eycken, Nigel Olisa, Eduard Vieta, Natalia E Fares-Otero, Thomas G Schulze, Therese van Amelsvoort, Bea Campforts, Emma de Brabander, Moritz J Rossner, Sven P Wichert, Roos van Westrhenen

Personal recovery is an increasingly used outcome for evaluating the effect of interventions on mental health disorders like affective and anxiety disorders, emphasizing living a meaningful life despite psychiatric challenges. The Recovery Assessment Scale (RAS) is a widely used measure of personal recovery with strong psychometric properties, but evidence for its sensitivity to change is limited. We conducted a systematic review and meta-analysis of 27 randomized controlled trials, found through systematic search on December 8th, 2022, which was updated in March 2024, to evaluate whether the RAS is sensitive to measuring change after interventions. Among the included randomized controlled trials, several focused on affective disorders, allowing us to examine the RAS in diagnostic groups directly relevant to affective disorder research. Results showed that participants in intervention groups demonstrated significantly greater pre-post changes in RAS scores compared to control groups for total RAS scores, with a small overall effect size (hedges g = 0.249). This effect was moderated by participants' diagnoses, but not by the specific RAS version, length of follow-up period, significant clinical outcome measured or baseline RAS scores. No significant findings were found for subdomain scores. Our results demonstrate that the RAS can detect change in a randomized controlled setting, although this design cannot fully isolate intervention effects from the instrument's inherent responsiveness. A better understanding of the scale's ability to capture changes in personal recovery can inform the development of more effective, personalized interventions, ultimately improving the well-being and quality of life of individuals with psychiatric disorders.

Copyright © 2026. Published by Elsevier B.V.

PMID: 41620183

Participating cluster members