Mindfulness-Based Cognitive Therapy (MBCT) for the Prevention of Depression Relapse or Recurrence Using Videoconference

Objectives: To evaluate the clinical- and cost-effectiveness of two delivery formats of mindfulness-based cognitive therapy (MBCT) - (i) MBCT delivered through eight videoconference group sessions (group MBCT), and (ii) MBCT delivered through a self-help manual plus three low-intensity videoconference group support sessions (supported MBCT self-help) - when added to treatment as usual (TAU; mental health outpatient care), compared with TAU alone, in patients with recurrent depression, either in an acute episode or in remission. We will identify potential predictors and moderators of response and examine psychological mechanisms of change. Study design: Multicentre randomised controlled trial, with pre-post, 6- and 12-months follow-ups. Centres: Parc Sanitari Sant Joan de Déu (St. Boi de Llobregat), Hospital Universitario Miguel Servet (Zaragoza), Hospital del Mar (Barcelona), Fundació Centre Higiene Mental Nou Barris (Barcelona), and Hospital General Universitario de Elche (Elche, Alicante). Participants: Adult patients (n=342) with recurrent depression, either in an acute episode or in remission, will be randomly assigned to group MBCT, supported MBCT self-help, or TAU. Primary outcome: transitions between clinically defined states of recurrent depression over the 12-month follow-up (DSM-5). The primary confirmatory transition is from remission to a major depressive episode (relapse/recurrence), whereas the transition from acute depression to remission will be examined as an exploratory objective. Secondary outcomes: depressive symptom severity, negative affect, mental well-being, quality of life, costs, and quality-adjusted life years. Process measures: mindfulness, positive affect, and self-compassion. Main statistical analyses: The primary analysis will use the intention-to-treat (ITT) principle and continuous-time multi-state modelling to estimate transition-specific hazard ratios. Secondary outcomes, predictors, and moderators will be analysed using regression