The clinical effectiveness of using a predictive algorithm to guide antidepressant treatment in primary care (PReDicT): an open-label, randomised controlled trial

Michael Browning, Amy C. Bilderbeck, Rebecca Dias, Colin T. Dourish, Jonathan Kingslake, Jürgen Deckert, Guy M. Goodwin, Philip Gorwood, Boliang Guo, Catherine J. Harmer, Richard Morriss, Andreas Reif, Henricus G. Ruhe, Anneke van Schaik, Judit Simon, Victor Perez Sola, Dick J. Veltman, Matilde Elices, Anne G. Lever, Andreas MenkeElisabetta Scanferla, Michael Stäblein, Gerard R. Dawson

Research output: Contribution to journalArticleAcademicpeer-review

31 Citations (Scopus)

Abstract

Depressed patients often do not respond to the first antidepressant prescribed, resulting in sequential trials of different medications. Personalised medicine offers a means of reducing this delay; however, the clinical effectiveness of personalised approaches to antidepressant treatment has not previously been tested. We assessed the clinical effectiveness of using a predictive algorithm, based on behavioural tests of affective cognition and subjective symptoms, to guide antidepressant treatment. We conducted a multicentre, open-label, randomised controlled trial in 913 medication-free depressed patients. Patients were randomly assigned to have their antidepressant treatment guided by a predictive algorithm or treatment as usual (TaU). The primary outcome was the response of depression symptoms, defined as a 50% or greater reduction in baseline score of the QIDS-SR-16 scale, at week 8. Additional prespecified outcomes included symptoms of anxiety at week 8, and symptoms of depression and functional outcome at weeks 8, 24 and 48. The response rate of depressive symptoms at week 8 in the PReDicT (55.9%) and TaU (51.8%) arms did not differ significantly (odds ratio: 1.18 (95% CI: 0.89–1.56), P = 0.25). However, there was a significantly greater reduction of anxiety in week 8 and a greater improvement in functional outcome at week 24 in the PReDicT arm. Use of the PReDicT test did not increase the rate of response to antidepressant treatment estimated by depressive symptoms but did improve symptoms of anxiety at week 8 and functional outcome at week 24. Our findings indicate that personalisation of antidepressant treatment may improve outcomes in depressed patients.

Original languageEnglish
Pages (from-to)1307-1314
Number of pages8
JournalNeuropsychopharmacology
Volume46
Issue number7
DOIs
Publication statusPublished - Jun 2021

Cite this