Depression follow-up monitoring with the PHQ-9: open cluster-randomised controlled trial

Lisa de Beste; Ieteke Ossendrijver; Lukas Koet; Madelon van Tilborg-den Boeft

doi:10.3399/bjgp24X737109

With great interest we read ‘Depression follow-up monitoring with the PHQ-9: open cluster-randomised controlled trial’ by Kendrick et al.¹ In this study the researchers investigated whether the use of the PHQ-9 questionnaire, a patient-reported outcome measure, improves the outcomes of patients with depressive problems in general practices. The authors conclude that there is no evidence for improved depression outcomes when the PHQ-9 is used. We thank the authors for their extensive work on this important topic. The utilised study design is appropriate to answer the research question and the results are presented clearly. However, after reading this article, several questions remain unanswered to us.

First, a high withdrawal rate from practices was observed, alongside a low participation rate of the patients. Out of 11 468 patients approached, only 529 eventually consented. What are the possible explanations for these low participation rates? The differences in response rate and in baseline depression between the intervention and control group suggest that participation bias may be present. Did the authors assess whether the participating patients are representative of the overall population? Second, many of the presented results are based on post-hoc analyses. Although the authors were transparent about this, it remains unclear whether the presented post-hoc analyses included all or a subset of those performed. Did the authors report all the post-hoc analyses they performed? Post-hoc analyses are known to increase the risk of type-1 errors and can lead to data-driven hypothesis generation.

Last, it appears that essential background information is missing. It would have been helpful to have more information about the PHQ-9 questionnaire, its purpose, and the presumed mechanism of action. How were the GPs taking these scores into account when managing the patients? Besides that, it would have been beneficial to have more information regarding the usual care provided to the control group.

We appreciate the authors’ effort to conduct this study. However, we seek additional clarifying details and information.

Thank you for taking our letter into consideration. We look forward to your response.

Reference

1.↵
1. Kendrick T,
2. Dowrick C,
3. Lewis G,
4. et al.
(2024) Depression follow-up monitoring with the PHQ-9: open cluster-randomised controlled trial. Br J Gen Pract, DOI: https://doi.org/10.3399/BJGP.2023.0539.