Attrition in treatment-resistant depression: predictors and clinical impact. (July 2019)

• Record Type: Journal Article
• Authors: Olgiati, Paolo; Serretti, Alessandro; Souery, Daniel; Kasper, Siegfried; Kraus, Christoph; Montgomery, Stuart; Zohar, Joseph; Mendlewicz, Julien
• Is Part Of: International clinical psychopharmacology. Volume 34:Number 4(2019:Jul.)
• Publication Date: 2019-07
• Subjects: attrition -- antidepressant -- improvement -- predictor -- treatment-resistant depression; Psychopharmacology -- Periodicals; Psychotropic drugs -- Periodicals; Psychopharmacology -- periodicals; Psychotropic Drugs -- periodicals; 615.7805
• DOI: 10.1097/YIC.0000000000000261
• Languages: English
• ISSNs: 0268-1315

Abstract

Abstract : The aim of this study was to investigate attrition (dropout) during a second antidepressant trial in treatment-resistant depression. Three hundred forty-two outpatients with major depressive disorder and lack of response to a prior antidepressant were treated with venlafaxine for 6 weeks. Sociodemographic and clinical characteristics were compared between the attrition and non-attrition groups. Attrition was reported in 65 patients (19%), of whom 30 patients (46%) dropped out within week 4. The characteristics of dropout patients included a longer duration of depressive episode ( P = 0.011) and lower antidepressant doses ( P < 0.0001) as a consequence of a faster decrease (week 2) in depressive symptoms ( P = 0.028). However, by controlling for early improvement, dropout subjects were associated with a smaller probability of antidepressant response (odds ratio = 0.16▪.83). A decrease of at least 30% in Montgomery Asberg Depression Rating Scale on day 14 predicted subsequent dropout with high specificity (81.9%▪1.0%) but lower sensitivity (19.6%▪2.8%) for clinical use. Patients who have been depressed for a longer period and show an initial improvement of symptoms after changing their antidepressant may be at increased risk for drop out. Further studies are necessary to ascertain the usefulness of these characteristics for predicting attrition.