Internet and telephone support for discontinuing long-term antidepressants: cluster randomized trial REDUCE open pragmatic effectiveness trial in UK primary care

Importance There is significant concern regarding increasing long-term antidepressant treatment for depression, beyond an evidence-based duration. Objective 2 To determine whether adding Internet and telephone support, to family practitioner review to consider discontinuing long-term antidepressant treatment, is safe and more effective than practitioner review alone. Design Cluster-randomized controlled trial, between 2019 and 2023, with remote computerized allocation and 12 months follow-up. Participants and researchers were aware of allocation, but analysis was blind.Setting 131 UK family practices. Participants Adults receiving antidepressants for >1 year for a first episode of depression, or >2 years for recurrent depression, currently well enough to consider discontinuation, and at low risk of relapse. Of 6,725 patients mailed invitations, 330 (4.9%) were eligible and consented. Interventions Internet and telephone self-management support, co-designed and co-produced with patients and practitioners. Outcomes Primary (safety) outcome: depression at six months (pre-specified complete cases analysis), testing for non-inferiority of the intervention within 2 points on the PHQ-9. Secondary outcomes (testing for superiority): antidepressant discontinuation, anxiety, quality of life, antidepressant withdrawal symptoms, mental wellbeing, enablement, satisfaction, use of services, and adverse events. Results Of 330 participants recruited (223 (68.6%) women; mean age 54.4 years; 179 in intervention practices and 151 in controls), 276 (83.6%) were followed-up at 6 months, and 240 (72.7%) at 12 months. The intervention proved non-inferior; in fact mean PHQ-9 scores were slightly lower in the intervention arm at 6 months in the complete cases analysis (4∙0 versus 5∙0; adjusted difference 1∙08; 95%CI -2∙06, -0∙09; p=0∙03), but not significantly different in an ITT multiple imputation sensitivity analysis (adjusted difference -0∙89 (95% CI -1∙90, 0∙11), p=0∙08). By six months antidepressants had been discontinued by 66/145 intervention arm participants 3 (45.5%) and 54/129 controls (41.9%) (adjusted odds ratio 1∙02; 95% CI 0∙52, 1∙99; p=0∙96). Antidepressant withdrawal symptoms and mental wellbeing were better in the intervention arm, but again differences were small. There were no significant differences in the other outcomes; 15% of participants in each arm experienced adverse events. Conclusions In this randomized controlled trial of adding Internet and telephone support to practitioner review for possible antidepressant discontinuation, depression was slightly better with support, but it did not significantly increase discontinuation. Improvements in antidepressant withdrawal symptoms and mental wellbeing were also small. There were no significant harms. Family practitioner review for possible discontinuation is safe and effective for more than 40% of patients willing to discontinue.