Antidepressant medication to prevent depression relapse in primary care: the ANTLER RCT
Larisa Duffy,Caroline S Clarke,Gemma Lewis,Louise Marston,Nick Freemantle,Simon Gilbody,Rachael Hunter,Tony Kendrick,David Kessler,Michael King,Paul Lanham,Dee Mangin,Michael Moore,Irwin Nazareth,Nicola Wiles,Faye Bacon,Molly Bird,Sally Brabyn,Alison Burns,Yvonne Donkor,Anna Hunt,Jodi Pervin,Glyn Lewis
DOI: https://doi.org/10.3310/hta25690
Abstract:Background: There has been a steady increase in the number of primary care patients receiving long-term maintenance antidepressant treatment, despite limited evidence of a benefit of this treatment beyond 8 months. Objective: The ANTidepressants to prevent reLapse in dEpRession (ANTLER) trial investigated the clinical effectiveness and cost-effectiveness of antidepressant medication in preventing relapse in UK primary care. Design: This was a Phase IV, double-blind, pragmatic, multisite, individually randomised parallel-group controlled trial, with follow-up at 6, 12, 26, 39 and 52 weeks. Participants were randomised using minimisation on centre, type of antidepressant and baseline depressive symptom score above or below the median using Clinical Interview Schedule - Revised (two categories). Statisticians were blind to allocation for the outcome analyses. Setting: General practices in London, Bristol, Southampton and York. Participants: Individuals aged 18-74 years who had experienced at least two episodes of depression and had been taking antidepressants for ≥ 9 months but felt well enough to consider stopping their medication. Those who met an International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, diagnosis of depression or with other psychiatric conditions were excluded. Intervention: At baseline, participants were taking citalopram 20 mg, sertraline 100 mg, fluoxetine 20 mg or mirtazapine 30 mg. They were randomised to either remain on their current medication or discontinue medication after a tapering period. Main outcome measures: The primary outcome was the time, in weeks, to the beginning of the first depressive episode after randomisation. This was measured by a retrospective Clinical Interview Schedule - Revised that assessed the onset of a depressive episode in the previous 12 weeks, and was conducted at 12, 26, 39 and 52 weeks. The depression-related resource use was collected over 12 months from medical records and patient-completed questionnaires. Quality-adjusted life-years were calculated using the EuroQol-5 Dimensions, five-level version. Results: Between 9 March 2017 and 1 March 2019, we randomised 238 participants to antidepressant continuation (the maintenance group) and 240 participants to antidepressant discontinuation (the discontinuation group). The time to relapse of depression was shorter in the discontinuation group, with a hazard ratio of 2.06 (95% confidence interval 1.56 to 2.70; p < 0.0001). By 52 weeks, relapse was experienced by 39% of those who continued antidepressants and 56% of those who discontinued antidepressants. The secondary analysis revealed that people who discontinued experienced more withdrawal symptoms than those who remained on medication, with the largest difference at 12 weeks. In the discontinuation group, 37% (95% confidence interval 28% to 45%) of participants remained on their randomised medication until the end of the trial. In total, 39% (95% confidence interval 32% to 45%) of participants in the discontinuation group returned to their original antidepressant compared with 20% (95% confidence interval 15% to 25%) of participants in maintenance group. The health economic evaluation demonstrated that participants randomised to discontinuation had worse utility scores at 3 months (-0.037, 95% confidence interval -0.059 to -0.015) and fewer quality-adjusted life-years over 12 months (-0.019, 95% confidence interval -0.035 to -0.003) than those randomised to continuation. The discontinuation pathway, besides giving worse outcomes, also cost more [extra £2.71 per patient over 12 months (95% confidence interval -£36.10 to £37.07)] than the continuation pathway, although the cost difference was not significant. Conclusions: Patients who discontinue long-term maintenance antidepressants in primary care are at increased risk of relapse and withdrawal symptoms. However, a substantial proportion of patients can discontinue antidepressants without relapse. Our findings will give patients and clinicians an estimate of the likely benefits and harms of stopping long-term maintenance antidepressants and improve shared decision-making. The participants may not have been representative of all people on long-term maintenance treatment and we could study only a restricted range of antidepressants and doses. Identifying patients who will not relapse if they discontinued antidepressants would be clinically important. Trial registration: Current Controlled Trials ISRCTN15969819 and EudraCT 2015-004210-26. Funding: This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 25, No. 69. See the NIHR Journals Library website for further project information.