A double blind placebo-controlled randomised trial of the addition of the antidepressant mirtazapine for patients with depression in primary care who have not responded to at least 6 weeks of antidepressant treatment

Design: A two parallel group multi-centre pragmatic placebo controlled randomised trial with allocation at the level of the individual. The primary outcome will be at 12 weeks, with further follow up at 6 and 12 months. Setting: Primary Care in Bristol, Exeter, Manchester and York Target population: Adult primary care patients who meet criteria for depression after at least 6 weeks treatment with either an SSRI or an SNRI antidepressant, taken at an adequate dose. Health technologies being assessed: The addition of mirtazapine to an SSRI or SNRI for the treatment of resistant depression. There is the potential for a synergistic action between SSRI/SNRI antidepressants and mirtazapine. There is evidence from previous studies that this could enhance clinical response compared to those patients receiving only an SSRI or SNRI. Mirtazapine is now off patent and inexpensive. Measurement of cost and outcomes: Primary outcome: Score on the Beck Depression Inventory (BDI) 12 weeks after randomisation, measured as a continuous variable. Secondary outcomes: BDI score at 24 weeks and 12 months (continuous and binary) and at all three time points, quality of life (EQ-5D), anxiety (GAD7) and adverse effects (ADEC). Adherence to medication will be measured using a brief questionnaire. We will conduct a cost-effectiveness analysis relating the costs of each strategy to the change in BDI scores at 12 weeks and 12 months; a cost-utility analysis relating the costs of each strategy to QALYs gained, using the EQ-5D-5L, at 12 months; and a cost consequences study at 12 months. Sample size: We estimate that a clinically important difference corresponds to about 3-4 points (0.35 standard deviations) on the BDI. With 200 participants in each group, we will have 91% power to detect a difference of 0.33 standard deviations at the 5% level. Allowing for 15% loss to follow-up at 12 weeks, we will need to recruit 472 patients. To detect a 50% reduction in symptoms, 200 patients in each group would yield 90% power to detect a difference between 30% and 46% response, or an odds ratio of 2, at a 2-sided 5% significance level Project timetable (including recruitment rate): We have recently completed the COBALT study, a trial of CBT for treatment resistant depression, randomising 469 patients over 18 months at 8.4 patients per centre per month. The population to be sampled is identical. We plan to recruit 470 patients over 18 months from 96 practices in 4 centres, a target of 7 patients randomised per month Year 1: Study set up. Months 7-12: recruitment begins for both trial and qualitative study. Follow up for primary outcome begins Year 2: recruitment continues for 12 months. Follow up for outcomes at all three time points throughout the year. Recruitment ends Years 3 &4: complete follow up: analysis, writing up and dissemination