Antidepressant medication to prevent depression relapse in primary care: the ANTLER RCT

Article history

The research reported in this issue of the journal was funded by the HTA programme as project number 13/115/48. The contractual start date was in May 2016. The draft report began editorial review in November 2020 and was accepted for publication in July 2021. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HTA editors and publisher have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this report.

Copyright statement

Copyright © 2021 Duffy et al. This work was produced by Duffy et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, distribution, reproduction and adaption in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.

2021 Duffy et al.

Description of the health problem

Depression is a leading cause of disability, with more than 300 million people having depressive illness worldwide. 1 According to the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10), depression is defined as a state of mental health that can be characterised by symptoms of low mood, loss of interest or pleasure, tearfulness, feelings of guilt or low self-worth, social withdrawal, disturbed sleep or appetite, poor concentration, tiredness and diminished activity. 2

Owing to difficulties in distinguishing between clinically significant and ‘normal’ mood changes, it is now generally accepted that depressive symptoms are a continuum of severity. 3 The severity of depressive illness can range from mild to severe. Even mild to moderate depression can impair people’s ability to function and cope with daily life. The diagnosis of depression is based not only on severity of symptoms but also on their duration and the degree of social and functional impairment. Depression is no longer thought of as a time-limited disorder with complete recovery after 4–6 months of treatment; it is now accepted that depression is often chronic (lasting for ≥ 2 years) or recurrent. Although it has been accepted that depression exists as a continuum, researchers and clinicians still use terms such as episode, relapse and chronicity to guide the diagnosis, and to inform and monitor treatment.

Treatment options for depression

People with depressive symptoms are mainly treated in primary care, and antidepressants are usually the first-line treatment; they are used to treat acute depressive symptoms and for maintenance treatment, that is to prevent relapse once an individual has recovered. The number of prescriptions for antidepressant medication has risen dramatically in high-income countries in recent decades, largely because of an increase in the number of patients receiving long-term treatment. 4,5 Psychological treatments, such as cognitive–behavioural therapy, are also effective treatments for depression, including for those who have not responded to antidepressants. 6

Economic consequences of depression

Depression not only causes marked emotional distress and interferes with daily function for the individual, but also has substantial negative social and financial impact on the wider community. 7 Every year, it has been estimated that depression reduces England’s national income (gross national product) by over 4% (approximately £80M). This reduction results from increased unemployment, a larger number of sick days and reduced productivity. It is also accompanied by increased welfare expenditure. 8,9

Current evidence on the effectiveness of maintenance treatment

A substantial proportion of the burden of depression arises from relapses, recurrence and chronicity. However, in contrast to the large number of drug trials in acute depression,10 there are relatively few studies assessing antidepressant efficacy in preventing relapse of depression during maintenance treatment. A number of systematic reviews and meta-analyses11–14 report a reduced risk of relapse rate in patients receiving antidepressant medication by between 50% and 70% compared with patients receiving placebo. The constituent studies had several limitations. These studies recruited patients from secondary care during an acute depressive episode, treated the patients with an open-label antidepressant and reported that only those who met criteria for recovery were eligible for randomisation to either remain on double-blind antidepressant or switch to placebo. Most studies were conducted during the 1980s or early 1990s in secondary care by pharmaceutical companies for regulatory purposes, using tricyclic antidepressants that are no longer widely used for depression. Most studies had either short (≤ 2 months) or intermediate (3–5 months) pre-randomisation treatment. It is difficult to generalise these findings to people currently receiving antidepressants in primary care, many of whom have been on maintenance treatment for some years. 15,16 Studies of the effectiveness of maintenance treatment for patients receiving antidepressants for longer than 8 months are rare and have several limitations. 17–19 All were very small (n < 20 participants in total) and had a poor follow-up rate.

NICE guidance on the treatment and management of depression in adults20 recommends that patients who have recovered from an episode of depression should stay on their antidepressant medication for at least 6 months after remission. The guidelines also suggest that the medication should be continued for 2 years after remission in those ‘at risk of relapse’; the risk is defined as ‘two or more episodes, residual symptoms or severe or prolonged episodes’ (© NICE 2010. Depression: Management of Depression in Primary and Secondary Care. Available from www.nice.org.uk/guidance/cg23. All rights reserved. Subject to Notice of rights. NICE guidance is prepared for the National Health Service in England. All NICE guidance is subject to regular review and may be updated or withdrawn. NICE accepts no responsibility for the use of its content in this product/publication). 20 However, there is no evidence that these proposed factors (number of episodes, residual symptoms and severity of previous episodes) affected the difference between antidepressant and placebo maintenance treatment. 13 NICE recommends that antidepressant maintenance treatments should continue to be used for 2 years for those at risk of relapse; however, NICE also recognise the uncertainty about the benefit of long-term maintenance treatment and recommend further research into its psychological and pharmacological effects.

At present, there is little evidence to support the use of long-term maintenance antidepressant treatment, despite its widespread use. Given the costs of treatment and medical supervision, the side effects of the medication and patients’ wish to be medication free, it is important to investigate this question further.

Health economic considerations are discussed and presented in Chapter 4.

Current evidence on withdrawal symptoms

There is also uncertainty about the frequency and severity of withdrawal symptoms after antidepressants are discontinued. Randomised, double-blind, placebo-controlled trials are required to provide robust evidence on withdrawal symptoms. A recent systematic review of such trials found evidence of withdrawal symptoms after antidepressant discontinuation, but the studies were too heterogeneous to establish the frequency, severity and timing of withdrawal effects. Almost all of these trials were rated as being of low quality because of the high risk of selection bias, attrition and incomplete outcome data. 21 Withdrawal symptoms are an important risk to evaluate when considering stopping maintenance treatment. They raise an important clinical question of the benefits of antidepressants compared with the possible risks.

Aim and objectives

The ANTLER trial was funded by the National Institute for Health Research (NIHR) as part of its Health Technology Assessment (HTA) programme. The overall aim of the ANTLER trial was to answer the following research question: ‘What is the clinical effectiveness and cost-effectiveness in UK primary care of continuing on long-term maintenance antidepressants compared with a placebo in preventing relapse of depression in those who have taken antidepressants for more than 9 months and who are now well enough to consider stopping maintenance treatment?’.

The trial was pragmatic, embedded in primary care and had broad inclusion criteria to increase the generalisability to the population currently receiving maintenance antidepressants.

Trial design

The ANTLER trial was a Phase IV, double-blind, pragmatic, multisite, individually randomised parallel-group controlled trial that was funded by the NIHR HTA programme. We recruited primary care patients who were taking one of four of the most commonly used antidepressant medications. At the point of recruitment, the patients were well enough to consider stopping their medication. Participants were recruited from primary care practices in four UK sites: London, Bristol, Southampton and York.

The trial compared maintenance treatment with antidepressant (citalopram 20 mg, sertraline 100 mg, fluoxetine 20 mg or mirtazapine 30 mg) treatment by replacing the medication with an identical placebo after a tapering period. We chose these antidepressant doses because they are the most common for long-term maintenance treatment in UK primary care and they simplified the manufacture of placebo and the conduct of the trial. In addition, there is no evidence that higher doses increase effectiveness. 22 The trial intervention was for 52 weeks and the participants were followed up at 6, 12, 26, 39 and 52 weeks.

Ethics approval and research governance

Ethics approval was obtained from the National Research Ethics Service (NRES) committee, East of England – Cambridge South (reference 16/EE/0032) and the Health Research Authority. A number of subsequent communications were sent to both the NRES and the Health Research Authority either seeking approval for substantial amendments or informing committees of minor changes. Clinical trial authorisation was given by Medicines and Healthcare products Regulatory Agency. The trial sponsor was University College London.

The ANTLER trial was registered on the Current Controlled Trials International Standard Randomised Controlled Trial Number (ISRCTN) registry (ISRCTN15969819; 21 September 2015) and also received a EudraCT number (2015-004210-26). As part of the NIHR Evaluation, Trials and Studies Co-ordinating Centre research portfolio, the trial was adopted and listed on the portfolio.

Participants

The trial recruited patients from 150 general practices across four research sites (London, Bristol, Southampton and York). Table 1 provides a summary of the main characteristics of the participating practices.

Recruitment

Recruitment started in March 2017. Within 2 years, 478 participants were recruited from general practices across our four research centres using two methods: record search and in-consultation recruitment. Figure 1 outlines a detailed flow chart of the stages of recruitment.

FIGURE 1.

Flow chart outlining the stages of recruitment.

Intervention

Subsequent assessments

The follow-up assessments were carried out at 6, 12, 26, 39 and 52 weeks after randomisation. Participants were invited to follow-up appointments unless they had withdrawn from the trial. Participants were followed up even if they stopped taking the trial medication. The follow-up appointments took place at the participant’s home, their general practice or university premises. Figure 2 describes the baseline and follow-up assessments. The baseline questionnaire can be found in Report Supplementary Material 2. The 6-week follow-up questionnaire can be found in Report Supplementary Material 3. The 12-, 26-, 39- and 52-week follow-up questionnaire can be found in Report Supplementary Material 4. The exit questionnaire can be found in Report Supplementary Material 7.

FIGURE 2.

Schedule of assessments. DESS, Discontinuation-Emergent Signs and Symptoms; EQ-5D-5L, EuroQol-5 Dimensions, five-level version; GAD-7, Generalised Anxiety Disorder-7; PHQ-9, Patient Health Questionnaire-9 items; SF-12, Short Form questionnaire-12 items.

Outcomes

Sample size

The sample size estimation was based on the evidence from systematic reviews available when the trial was designed. The reduction in the odds of relapse in an active group compared with a placebo group was estimated to be 70% in a systematic review by Geddes et al. ,11 65% by Kaymaz et al. 13 and Glue et al. 12 and 50% by NICE. 20 Between 15% and 22% of those taking the active drug relapsed in 12 months. To detect the difference between relapse rates of 15% (maintenance group) and 30% (discontinuation group) (hazard ratio 0.46), or between relapse rates of 20% (maintenance group) and 35% (discontinuation group) (hazard ratio 0.52), we estimated that the required sample sizes were 333 and 383 participants, respectively, for 90% power at the 5% significance level. Allowing for 20% attrition, we therefore proposed to recruit 479 participants. 38 Analyses are expressed with the discontinuation group as reference, equating to a hazard ratio of 1.92 for the power calculation.

Randomisation and blinding

Following completion of the baseline assessment, eligible participants who consented were randomised using the automated randomisation service provided by Sealed Envelope (London, UK; https://sealedenvelope.com). The randomisation was minimised by the four study centres, the four medications and the severity of depressive symptoms at baseline (two categories measured using the CIS-R). The dispensing pharmacy (University Hospitals Bristol Pharmacy) was informed of the randomised allocation and posted the medication by recorded delivery to either the participant’s home or the GP surgery at 8-week intervals. Trial participants, clinicians and all members of the research team were blinded to the trial treatment allocation. Statisticians analysed the data blind to allocation. Health economists were aware of the allocation so that they could cost the trial medications. Participants were free to withdraw from the medication at any time.

Together with trial medication, participants were posted a contact card so that any treating clinician could be unblinded to treatment allocation in case of a medical emergency (emergency unblinding) or to enable treatment decisions (early unblinding). If unblinding was required, a formal request by a clinician was made to the trial pharmacy (through the 24-hour contact number provided on the contact card) that had a list of the participants’ treatment allocations. The treating physician managed the medical emergency as appropriate on receipt of the treatment allocation.

The researcher recorded any breaking of the code and the reasons for doing so in the unblinding log. When possible, members of the research team remained blinded. Those participants who did not require emergency or early unblinding were unblinded on completion of the trial (routine unblinding). This information was provided to their GP by the pharmacy; the participant was encouraged to see their GP to discuss any further treatment during that consultation. The trial team remained blind to this information.

Statistical methods

Reliability of the retrospective Clinical Interview Schedule – Revised

We assessed the onset of a depressive episode and called this assessment rCIS-R and conducted a test–retest reliability study of rCIS-R. The study was nested within the ANTLER trial. The ANTLER trial participants were asked to complete the rCIS-R twice: at the beginning of one of the face-to-face follow-up appointments and at the end of the same appointment.

Results of the reliability study of the retrospective Clinical Interview Schedule – Revised

Of 478 participants who were recruited to the trial, 396 completed the rCIS-R twice. Two participants completed the rCIS-R at the 12-week follow-up appointment, 335 participants at 26 weeks, 42 participants at 39 weeks and 17 participants at 52 weeks. There were 106 male participants, the mean age was 55 years (SD 6 years) and 6% of the participants reported being from an ethnic minority group. The full description of the reliability study sample characteristics compared with the trial sample is in Table 3.

TABLE 3 - Baseline characteristics of the sample who completed the rCIS-R twice at the same follow-up appointment

A Level, Advanced Level; GCSE, General Certificate of Secondary Education.

Characteristic	Reliability study sample	Trial sample
Age (years), mean (SD)	55 (6)	54 (6)
Male, n/N (%)	106/396 (27)	128/478 (27)
Ethnicity, n/N (%)
White	373/396 (94)	449/473 (95)
Not white	23/396 (6)	25/473 (5)
Highest educational qualification, n/N (%)
Degree/higher degree	146/396 (37)	179/472 (38)
Diploma/A Levels or equivalent	127/396 (32)	148/472 (31)
GCSE or equivalent/other/none	123/396 (31)	145/472 (31)
Site, n/N (%)
London	170/396 (43)	199/478 (42)
Bristol	80/396 (20)	102/478 (21)
Southampton	84/396 (21)	96/478 (20)
York	62/396 (16)	81/478 (17)
Antidepressant, n/N (%)
Sertraline	62/396 (16)	78/478 (16)
Citalopram	183/396 (46)	223/478 (47)
Fluoxetine	135/396 (34)	160/478 (33)
Mirtazapine	16/396 (4)	17/478 (4)
CIS-R score at baseline, mean (SD)	5.0 (4.8)	5.1 (4.8)
PHQ-9 score at baseline, mean (SD)	3.8 (3.6)	3.8 (3.5)
Age first became aware of having depression (years), mean (SD)	32 (5)	32 (15)

The Cohen’s kappa for relapse in depression was 0.84 [95% confidence interval (CI) 0.71 to 0.97], which indicates excellent agreement between the first and the second completions of the rCIS-R (Tables 4 and 5). The level of agreement of the individual sections of the rCIS-R was also excellent (see Table 4). The agreement of the time of depression relapse was also assessed: both agreement of month κ = 0.84 (95% CI 0.71 to 0.97) and agreement of week κ = 0.87 (95% CI 0.74 to 1.00) of reappearance of depression.

TABLE 4 - Level of agreement of relapse in depression and individual symptoms between the first and the second completions of the rCIS-R

	Frequency present at the first completion (%)	Weighted Cohen’s kappa (95% CI)
Relapse	20	0.84 (0.71 to 0.97)
Symptoms
Depression or depressive mood	54	0.87 (0.77 to 0.97)
Depressive thoughts	50	0.87 (0.77 to 0.87)
Fatigue	56	0.85 (0.75 to 0.95)
Concentration	25	0.81 (0.72 to 0.91)
Sleep	52	0. 91 (0.82 to 1.00)

TABLE 5 - Table indicating agreement for relapse in depression between the first and the second completion of the rCIS-R

Relapsed on the second completion	Relapse on the completion (n)		Total (n)
	Did not relapse	Relapsed
Did not relapse	301	18	319
Relapsed	15	62	77
Total	316	80	396

The mean score for the first completion of the rCIS-R was 6.67 (SD 5.06) and the mean score for the second completion was 6.41 (SD 5.25). The percentage of participants meeting the relapse criteria at the first completion was 20% (n = 80) and at the second completion was 19% (n = 77). The mean total score difference was –0.25 (95% CI –0.43 to –0.07).

The Bland–Altman plot (Figure 3) shows the agreement between the first and the second completion of the rCIS-R. The intraclass correlation coefficient was 0.94 (95% CI 0.92 to 0.95).

FIGURE 3.

Difference in the total rCIS-R score against the mean total score.

We also looked at defining relapse in line with ICD-10 criteria: 18% of participants relapsed according to the ICD-10 diagnostic criteria for depression at the first (n = 72) and second (n = 70) completions. Cohen’s kappa for relapse of depression according to ICD-10 criteria was 0.74 (95% CI 0.64 to 0.84).

Public and patient involvement

Paul Lanham, our public and patient involvement (PPI) representative, is a co-applicant of the proposal and has been involved in all stages of the trial for almost 5 years. He is a former chairperson (1988–93) and director (1986–2009) of Depression Alliance [now merged with Mind (London, UK)] and provided input to previous studies on depression funded by the HTA programme (TREAD,52 CoBalT6 and PREVENT53).

In the development of the proposal, Paul Lanham was supportive of examination of the effectiveness of maintenance medication and wrote ‘From a user point of view this strikes me as being vitally important and I am delighted to be involved in it; I am sure that others (especially sufferers and GPs) will also welcome it and gain a great deal from its results’ (reproduced with permission from Paul Lanham, London, May 2014, personal communication). Paul is currently on maintenance antidepressants himself and added ‘I mistrust terms like “remission”, “well”, “normal” etc. People ask me if Citalopram helps; the answer is that I have no idea of what I would be like without it so I cannot judge. This [study] will hopefully determine whether such drugs are a help or not. It would be really valuable to know the answer to this through the study’ (reproduced with permission from Paul Lanham, personal communication). Paul was a co-applicant on the HTA proposal and a co-author on the final report.

During the trial, Paul Lanham was a member of the Trial Steering Group and the Trial Management Group, but he decided once the trial was established that his contribution was not required. In 2017, we recruited another PPI member, Lucy Carr, who was a former participant in a NIHR-funded trial, PANDA. 33 She was an independent member of the Trial Steering Committee and contributed, along with Paul Lanham, to the design and content of the study documentation, including patient information sheets and self-harm protocols.

In addition, we enlisted the support of the North London Service Users Research Forum (SURF). North London SURF was co-founded in 2007 by service users and clinical academic psychiatrists at University College London to provide meaningful consultation on research. It has 12 members who have mental health problems. Since 2007, it has been consulted on over 100 projects, and North London SURF members have also been invited to join steering/management groups on many of these. As a result, the group is very experienced and confident about the advice and input that it provides; members’ comments on the trial paperwork have been invaluable. The letter templates, patient information sheets and questionnaires were amended to reflect the North London SURF feedback. We also consulted on the protocol concerning self-harm or risk of self-harm, which was used if patients reported this in the course of the trial.

We also contacted Luke Montagu, who has experienced withdrawal symptoms from antidepressants and is a well-known activist in this area. Luke Montagu, along with others with lived experience, helped us modify the DESS scale measure of withdrawal symptoms. We shortened the scale (from 42 to 15 items) to improve its acceptability and selected the most commonly found items using published literature and a survey of Luke Montagu’s contacts. After consulting with the group, we included a question on electric sensations in the brain (brain zaps), apparently a very common, disabling symptom that has been poorly understood and not included in previous scales. Close involvement of the PPI for the duration of the trial has been invaluable for its success from the set-up stage onwards, shaping the study through to the discussion on the interpretation of results. We plan to carry on using the services users’ involvement in the design, documentation and analysis of any future studies.

Flow of participants in the trial

Recruitment began on 9 March 2017 and the last participant was randomised on 1 March 2019. The flow of participants through the trial is shown in the Consolidated Standards of Reporting Trials (CONSORT) flow diagram (Figure 4). The GP record search identified 23,429 potentially eligible patients, who were sent an invitation letter. Another 124 potentially eligible patients were referred during GP consultation, resulting in 1466 patients wanting to take part. Of these patients, 606 were eligible. A total of 478 participants were randomised: 238 to the maintenance group and 240 to the discontinuation group. All of the participants provided data on whether or not they relapsed; however, 10 participants (maintenance group, n = 6; discontinuation group, n = 4) did not provide data on timing of relapse, so could not be included in the analysis of the primary outcome.

FIGURE 4.

The CONSORT flow diagram for the ANTLER trial.

Baseline comparability

The baseline characteristics of the sample overall and by treatment group are shown in Table 6. The treatment groups were well balanced at baseline, with just over one-quarter of male participants and a mean age of 54 years (SD 13 years) in the maintenance group and 55 years (SD 12 years) in the discontinuation group. Just over 40% of participants were recruited from London, 20% from each of Bristol and Southampton, and 17% from York. Under half of the participants were taking citalopram, one-third fluoxetine, one-sixth sertraline and < 5% mirtazapine. Almost three-quarters of the participants had taken antidepressants for more than 3 years, with over one-third taking them for ≥ 6 years. The mean age at becoming aware of having depression was 33 years (SD 16 years) in the maintenance group and 32 years (SD 14 years) in the discontinuation group. The median time between randomisation and taking the trial medication was 9 days (interquartile range 6–13 days) in the maintenance group and 8 days (interquartile range 6–13 days) in the discontinuation group.

Primary outcome

The time to relapse was shorter in participants who discontinued antidepressants than in those who stayed on maintenance treatment [hazard ratio (HR) 2.06, 95% CI 1.56 to 2.70; p < 0.0001] (Table 7 and Figure 5). This result was unaltered after sensitivity analyses.

Overall, relapse was experienced by 39% (n = 92/238) (95% CI 32% to 45%) of participants in the maintenance group and 56% (n = 135/240) (95% CI 50% to 63%) of participants in the discontinuation group by the end of the trial (52 weeks). Relapses according to treatment group are shown in the tree diagrams (see Report Supplementary Material 9).

FIGURE 5.

Kaplan–Meier plot for the primary outcome. From N Engl J Med, Lewis G, Marston L, Duffy L, Freemantle N, Gilbody S, Hunter R, et al. , Maintenance or discontinuation of antidepressants in primary care, vol. 385, pp. 1275–67. 34 Copyright © 2021 Massachusetts Medical Society. Reprinted with permission.

From N Engl J Med, Lewis G, Marston L, Duffy L, Freemantle N, Gilbody S, Hunter R, et al. , Maintenance or discontinuation of antidepressants in primary care, vol. 385, pp. 1275–67. 34 Copyright © 2021 Massachusetts Medical Society. Reprinted with permission.

Secondary outcomes

The PHQ-9 and GAD-7 scores were higher (worse) in the discontinuation group than in the maintenance group at 12 and 26 weeks, the highest of which for both was at 12 weeks (PHQ-9: coefficient 2.16, 95% CI 1.47 to 2.84; GAD-7: coefficient 2.40, 95% CI 1.81 to 2.99).

The number of withdrawal symptoms reported was higher in the discontinuation group than in the maintenance group at 6, 12, 26 and 39 weeks, with the difference largest at 12 weeks (coefficient 1.87, 95% CI 1.46 to 2.28).

Participants in the discontinuation group had lower (worse) SF-12 mental health-related quality-of-life scores at 12, 26 and 39 weeks, with the difference largest at 12 weeks (coefficient –4.86, 95% CI –6.44 to –3.29). At 12 weeks, in the discontinuation group, the odds of feeling worse, determined using the Global Rating Question, was more than twice that (odds ratio 2.88, 95% CI 1.90 to 4.38) in the maintenance group (Table 8).

Missing data

The results of the predictors of missingness analysis are in Report Supplementary Material 12. We conducted sensitivity analyses that included predictors of missingness for continuous secondary outcomes. Results for all outcomes were similar to the main analyses (see Table 8) when including predictors of missingness in the models (Table 9).

Adherence to trial medication

A larger percentage of participants in the placebo than maintenance group stopped taking their trial medication before the end of the trial (48% vs. 30%; HR 2.26, 95% CI 1.67 to 3.07). By 52 weeks, 39% (95% CI 32% to 45%) of participants in the discontinuation group and 20% (95% CI 15% to 25%) in the maintenance group had stopped taking trial medication and had returned to an antidepressant prescribed by their GP.

Subgroup analyses

For the subgroup analyses for the primary and secondary outcomes, there was no evidence of any differences according to the number of previous episodes of depression dichotomised at two compared with three or more or types of antidepressant. The most consistent evidence that we found supported a larger difference between groups in those who were younger at onset. For example, the p-value for interaction between age and PHQ-9 at 12 weeks was 0.0001. Among participants who were older at the onset of depression, those in the discontinuation group were less likely to relapse, although this finding was not statistically significant (p = 0.0553) (Tables 10–13).

Post hoc analyses

Of the 134 participants randomised to the discontinuation group who had relapsed by the end of the trial, 49 (37%, 95% CI 28% to 45%) remained on trial medication, 71 (53%, 95% CI 44% to 62%) had returned to a known antidepressant and 14 (10%, 95% CI 6% to 17%) were not taking any antidepressant. Of the 89 participants randomised to the maintenance group who had relapsed, 46 (52%, 95% CI 41% to 62%) remained on trial medication, 32 (36%, 95% CI 26% to 47%) had returned to a known antidepressant and 11 (12%, 95% CI 6% to 21%) were not taking any antidepressant.

In total, 59% (141/240) of participants in the discontinuation group were unblinded through either withdrawal from the trial or emergency code break. The rate of unblinding was much lower in the maintenance group (29%; 68/236). Participants were asked whether they thought that they were taking the active drug (antidepressant) or the placebo. Over the course of the trial, 71% (162/228) of participants in the discontinuation group and 47% (108/232) of participants in the maintenance group correctly guessed their randomised group at some time before being unblinded.

We also investigated whether or not withdrawal symptoms differed according to antidepressant class. At 6 weeks, there was weak evidence (p = 0.07) that the effect of discontinuation (compared with maintenance) on withdrawal symptoms was smaller among those receiving fluoxetine than sertraline and citalopram. At 12 weeks, there was stronger evidence that the effect of discontinuation compared with maintenance on withdrawal symptoms differed by group (p = 0.002). Withdrawal symptoms were less common in those taking fluoxetine and citalopram than in those taking sertraline. There was no evidence of an interaction between treatment group and antidepressant class at week 26, 39 or 52. Withdrawal symptoms in those who discontinued and those who remained on maintenance antidepressant are shown in Report Supplementary Material 10, according to antidepressant class.

Introduction

The aim of the economic evaluation was to calculate the mean incremental cost per quality-adjusted life-year (QALY) gained by discontinuing antidepressant medication and replacing it with placebo, compared with antidepressant maintenance, from a health-care cost perspective, using trial data collected over 12 months from participants and primary care electronic medical records. SSRIs are the most common and recommended antidepressants, and their mean purchase cost is relatively low, around 4p per day. The majority of analyses evaluating the cost-effectiveness of prescribing antidepressants for depression have conducted head-to-head decision modelling of different antidepressants to determine the most cost-effective antidepressant to treat current symptoms, rather than considering the question of the wider cost-effectiveness of their long-term use,20,54,55 with analyses rarely going beyond a 12-month time horizon. The impact of side effects and withdrawal symptoms following long-term use is rarely given consideration; therefore, only limited, poor-quality data are available on which decision modelling could be based to describe long-term use. 21 In this chapter, we present the results of a trial-based cost–utility analysis (CUA) comparing discontinuation with continued antidepressant maintenance in primary care in England over 12 months, using patient-level data on health-care resource use and a preference-based measure of health-related quality of life [EuroQol-5 Dimensions, five-level version (EQ-5D-5L)].

Methods

Economic evaluation analytical methods

Descriptive statistics for the primary analysis, which used multiple imputation for missing utility scores and CSRI information, are reported for resource use, costs and utilities at each time point. The baseline age and SF-12 Physical Component Summary score were identified as predictors of missingness for the imputations.

The mean per-participant differences in 12-month costs and QALYs by treatment group were jointly estimated via bootstrapped seemingly unrelated regression, with 1000 iterations to account for the correlation between costs and QALYs,69 adjusting for baseline values and the minimisation variables of study centre, ANTLER trial medication and binary severity of depressive symptoms at baseline, with imputed data sets combined according to Rubin’s rules. 70 The primary economic analysis was calculated using the multiple imputation data set and the bootstrapped, seemingly unrelated, regression results, as set out by Leurent et al. 71

We took a probabilistic approach to aid decision-making for resource allocation and calculated the probability that discontinuing antidepressants to zero dose was cost-effective for a range of thresholds of cost per QALY gained compared with antidepressant maintenance. The ‘new’ treatment here is discontinuation and the ‘old’ treatment is maintenance; therefore, the incremental costs and QALYs are calculated as discontinuation pathway values minus maintenance pathway values. In the protocol paper72 and statistical analysis plan, we proposed to use the placebo as the ‘old’ treatment, as might be usual in most placebo-controlled trials. However, on reflection, we realised that the ‘new’ intervention in our participants was to discontinue antidepressants, so we decided to conduct the analysis from this viewpoint.

The incremental cost-effectiveness ratio (ICER) for each analysis was calculated as the mean estimated difference in costs divided by the mean estimated difference in QALYs. The bootstrapped results were plotted on cost-effectiveness planes (CEPs) and the proportions of estimates that were above the cost-effectiveness threshold were plotted on corresponding cost-effectiveness acceptability curves (CEACs) for a range of thresholds.

Health economics results

Utility scores and quality-adjusted life-years

Table 19 shows the complete-case raw utility scores at each time point and the QALYs calculated as the area under the curve, as well as the adjusted differences according to treatment group. The p-values indicate whether or not the difference in the value in that row (e.g. 3-month utility) is statistically significantly different between the two treatment groups, using linear regression. Table A8.14 in Report Supplementary Material 8 shows the complete-case raw and adjusted utility scores, and the adjusted differences according to both treatment group and relapse status. Table A8.15 in Report Supplementary Material 8 shows the mean adjusted differences in QALYs calculated over the 12-month period, and the mean adjusted difference in utility score at 3 months. These are given by both treatment group and relapse status in a post hoc analysis designed to explore whether or not relapse was driving the differences observed between the treatment groups.

TABLE 19 - Mean raw utilities at baseline and all follow-up points and mean 12-month QALYs, and estimated adjusted differences between the discontinuation group and the maintenance group

p < 0.05.

Time point	Treatment group				Discontinuation vs. maintenance
	Maintenance (0)		Discontinuation (1)
	n	Mean (raw) (SD)	n	Mean (raw) (SD)	Adjusted difference (95% CI)	p-value
EQ-5D-5L TTO: base case
Baseline	237	0.868 (0.151)	240	0.889 (0.114)
3 months	228	0.872 (0.146)	215	0.849 (0.145)	–0.037 (–0.059 to –0.015)	0.001a
6 months	210	0.875 (0.142)	191	0.870 (0.145)	–0.014 (–0.038 to 0.011)	0.268
9 months	212	0.872 (0.148)	183	0.882 (0.122)	–0.001 (–0.023 to 0.022)	0.953
12 months	210	0.879 (0.144)	181	0.871 (0.151)	–0.021 (–0.045 to 0.003)	0.090
QALYs	205	0.876 (0.119)	173	0.869 (0.115)	–0.019 (–0.035 to –0.003)	0.020a
EQ-5D-5L mapping: secondary analysis
Baseline	237	0.804 (0.186)	240	0.828 (0.146)
3 months	228	0.810 (0.178)	215	0.782 (0.183)	–0.044 (–0.072 to –0.017)	0.002a
6 months	210	0.815 (0.176)	191	0.808 (0.181)	–0.019 (–0.049 to 0.011)	0.218
9 months	212	0.809 (0.181)	183	0.824 (0.146)	0.003 (–0.024 to 0.031)	0.828
12 months	210	0.815 (0.181)	181	0.810 (0.183)	–0.019 (–0.049 to 0.011)	0.214
QALYs	205	0.814 (0.151)	173	0.806 (0.139)	–0.022 (–0.042 to –0.002)	0.028a
SF-12: secondary analysis
Baseline	237	0.754 (0.124)	239	0.774 (0.125)
3 months	227	0.745 (0.132)	216	0.712 (0.128)	–0.045 (–0.065 to –0.024)	< 0.001a
6 months	210	0.754 (0.131)	192	0.738 (0.134)	–0.025 (–0.049 to –0.001)	0.040a
9 months	212	0.764 (0.134)	183	0.759 (0.132)	–0.012 (–0.034 to 0.010)	0.287
12 months	210	0.766 (0.136)	181	0.749 (0.135)	–0.025 (–0.051 to 0.001)	0.055
QALYs	204	0.759 (0.103)	174	0.743 (0.100)	–0.027 (–0.043 to –0.011)	0.001a

Figure 6 shows the raw, unadjusted complete-case mean utility scores at each time point by treatment group for each of the three methods: EQ-5D-5L TTO (primary economic analysis), EQ-5D-5L mapping and SF-12/SF-6D. The dashed lines are the discontinuation group and the continuous lines are the maintenance group. The highest values are for EQ-5D-5L TTO (primary analysis) and the lowest values are for SF-12/SF-6D.

FIGURE 6.

Raw, unadjusted complete-case mean utility scores at each time point by treatment group for each of the three methods. Adapted with permission from Clarke et al. 75 This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. The figure includes minor additions and formatting changes to the original figure.

Adapted with permission from Clarke et al. 75 This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. The figure includes minor additions and formatting changes to the original figure.

Cost–utility analysis

The overall result of the cost–utility analysis is summarised as the ICER, which is the mean incremental cost per QALY gained of discontinuing antidepressant medication compared with maintenance antidepressant medication. In the primary economic analysis, with utilities and CSRI costs multiply imputed, discontinuation was dominated by maintenance in that the former cost more (£2.71, 95% CI –£36.10 to £37.07) and resulted in fewer QALYs (–0.019, 95% CI –0.035 to –0.003) than maintenance, although the 95% CI crosses zero in the costs. This means that the bootstrapped differences in costs and QALYs lie predominantly in the north-west quadrant of the CEP (Figure 7), characterised by the ‘new’ intervention under assessment, in this case discontinuation of antidepressants, incurring higher costs and providing fewer QALYs than the ‘old’ treatment, which in this case is maintenance of antidepressants. ICERs for the primary and secondary analyses are given in Tables 20 (primary analysis using health-care and social care perspective) and 21 (wider societal perspective).

FIGURE 7.

Cost-effectiveness plane using multiple imputation for utilities and CSRI costs, showing 1000 bootstrapped results from the seemingly unrelated regression of costs (NHS and PSS perspective) and QALYs (EQ-5D-5L TTO). Dark blue point is the mean costs vs. mean QALYs.

TABLE 20 - Summary of the bootstrapped incremental costs and QALYs, and the ICERs, using the health-care and social care cost perspective

MICE, multiple imputation by chained equations.

Quality-of-life instrument and value set	Difference (95% CI)		ICER
	Costs (£)	QALYs
Using MICE for missing utilities and CSRI costs
EQ-5D-5L TTO	2.71 (–36.10 to 37.07)	–0.010 (–0.024 to 0.004)	Maintenance dominates discontinuation
EQ-5D-5L mapping	4.19 (–31.42 to 36.40)	–0.012 (–0.028 to 0.004)	Maintenance dominates discontinuation
SF-12/SF-6D	2.16 (–36.58 to 36.41)	–0.020 (–0.033 to –0.008)	Maintenance dominates discontinuation
Using complete-case analysis only
EQ-5D-5L TTO	26.29 (–3.32 to 57.29)	–0.019 (–0.032 to –0.005)	Maintenance dominates discontinuation
EQ-5D-5L mapping	26.22 (–3.61 to 57.29)	–0.022 (–0.038 to –0.006)	Maintenance dominates discontinuation
SF-12/SF-6D	23.93 (–6.66 to 56.67)	–0.026 (–0.039 to –0.013)	Maintenance dominates discontinuation

TABLE 21 - Overall incremental costs and QALYs using the wider societal perspective

MICE, multiple imputation by chained equations.

Quality-of-life instrument and value set	Difference (95% CI)		ICER
	Costs (£)	QALYs
Using MICE for missing utilities and CSRI costs
EQ-5D-5L TTO	22.30 (–179.25 to 218.76)	–0.011 (–0.024 to 0.003)	Maintenance dominates discontinuation
EQ-5D-5L mapping	21.46 (–168.86 to 210.15)	–0.012 (–0.029 to 0.003)	Maintenance dominates discontinuation
SF-12/SF-6D	21.42 (–179.47 to 216.89)	–0.020 (–0.033 to –0.008)	Maintenance dominates discontinuation
Using complete-case analysis only
EQ-5D-5L TTO	105.99 (–128.37 to 331.22)	–0.019 (–0.032 to –0.006)	Maintenance dominates discontinuation
EQ-5D-5L mapping	106.02 (–128.38 to 331.39)	–0.022 (–0.039 to –0.006)	Maintenance dominates discontinuation
SF-12/SF-6D	101.48 (–142.45 to 341.27)	–0.026 (–0.040 to –0.013)	Maintenance dominates discontinuation

The information from the CEP was translated onto the CEAC (Figure 8), which shows the likelihood of discontinuation being cost-effective at a range of values of the cost-effectiveness threshold. Figure 8 shows values up to £10,000 per QALY and that at the standard QALY threshold values of £20,000 and £30,000 per QALY gained there was a 12.9% and 12.4% probability that discontinuation was cost-effective compared with maintenance, respectively. The CEAC lies below the 50% region for all thresholds, in agreement with the conclusion that discontinuation is dominated by maintenance.

FIGURE 8.

Cost-effectiveness acceptability curve generated from the CEP in Figure 7 showing the likelihood of the discontinuation group being sufficiently cost-effective compared with the maintenance group, at a range of values for the cost-effectiveness threshold from zero to £10,000 per QALY.

Summary of findings

Our large pragmatic trial of the effectiveness of maintenance antidepressants in primary care found that the rate of relapse was twice as high in those who received placebo as in those who continued to take antidepressant medication during 12 months of follow-up (hazard ratio 2.04, 95% CI 1.55 to 2.68). Most relapses occurred 12–26 weeks after randomisation and 4–18 weeks after medication was tapered.

Our analyses of secondary outcomes found strong evidence that depressive and anxiety symptoms at 12 weeks were higher in those receiving the placebo than in those receiving antidepressant medication, and the size of this difference was likely to be clinically meaningful. 76 Weaker evidence of a smaller difference in depressive and anxiety symptoms remained at 6 months, but attenuated thereafter. People in the discontinuation group were more likely than those in the maintenance group to self-report worsening of mood and had lower mental health-related quality of life at 12 weeks, although these differences were not evident at any other time point. At 12 weeks, 44% of those who discontinued their antidepressants reported feeling worse, compared with 21% of those who remained on their medication. Such global changes in mental health are used to calculate minimal clinically important differences and are regarded as patient-centred indicators of clinically meaningful change. 77 We also found strong evidence that patients who discontinued antidepressants stopped their trial medication sooner than those who remained on maintenance treatment (HR 2.26, 95% CI 1.67 to 3.07). By the end of the trial, 39% (95% CI 32% to 45%) of participants in the discontinuation group had returned to an active antidepressant prescribed by their clinician, which may explain why treatment effects attenuated after the 12-week follow-up.

People who discontinued their antidepressant experienced more withdrawal symptoms than those who remained on maintenance treatment at every time point except 52 weeks, by which time the difference had attenuated substantially (and many people in the discontinuation group had returned to taking an antidepressant). The difference in withdrawal symptoms between the groups was largest at 12 weeks and reduced as follow-up progressed. There was no evidence of a difference in physical symptoms that could be attributed to antidepressant side effects (in contrast to withdrawal effects) until 39 and 52 weeks after randomisation, when those receiving an antidepressant reported slightly more physical symptoms than those receiving placebo. It is possible that any difference in side effects was masked by the increase in depressive and/or withdrawal symptoms in the discontinuation group.

Summary of economic evaluation findings

The cost–utility analysis suggests that there is a low probability that discontinuing antidepressant medications with replacement by placebo is cost-effective compared with continued maintenance of antidepressant treatment in this population. Participants randomised to the discontinuation group experienced significantly fewer QALYs over 12 months than those in the maintenance group, with this difference primarily driven by an increased rate of relapse. Participants who were randomised to the discontinuation group also had higher GP consultation costs and Improving Access to Psychological Therapies costs. The difference in GP costs appeared to be driven by a shorter time to relapse in the discontinuation group, potentially because participants arranged to see their GP following relapse to review their medication, with 53% (95% CI 44% to 62%) of those who relapsed in the discontinuation group having returned to a known antidepressant before the end of the trial.

A limitation of the analysis is that the time horizon is the same as the follow-up period for the trial (i.e. only 12 months). It is possible that there are longer-term impacts on costs and QALYs, such as medication side effects, feelings of stability derived from maintenance medication or feelings of liberation derived from having become ‘medication free’, that are not captured by this analysis. However, decision analytic models of depression rarely go beyond 12 months, and we are not aware of any other studies with longer time horizons on which a decision model with a longer time horizon could be based; therefore, any modelling would have required many untested assumptions. In addition, given the relatively large difference in QALYs between tapering and maintenance compared with the relatively small difference in costs, it seems unlikely that discontinuation could potentially be cost-effective over the long term. Missing values also represent a limitation to the analysis; this analysis used multiple imputation, which carries the assumption that values are missing at random, but we cannot know whether or not this is correct. A further limitation of the analysis is that the effectiveness measures used were generic health-related quality-of-life measures and, although there is no evidence to suggest that other quality-of-life measures might perform better in this context, it is still possible that information regarding other factors that are important to participants was not captured. Finally, we compared maintenance antidepressants with the replacement of antidepressants with a placebo over a tapering period, although the use of placebo after discontinuation is not an indicated treatment. We chose to include a placebo to ensure that we were studying the pharmacological effects of the antidepressant drug, which is essential to inform policy. However, it is likely that the difference between maintenance and discontinuation without a placebo would be larger.

At a £20,000 to £30,000 threshold for a QALY gain (i.e. the preferred NICE threshold78) there would need to be a cost saving of £200–600 per person for tapering over the lifetime horizon following the first 12 months to balance the QALYs lost during the first 12-month period by discontinuing the medication; it is not clear where in the longer-term patient pathway this could potentially occur. This highlights the need for further research into the longer-term implications of discontinuing antidepressant use compared with continuing with long-term maintenance for the prevention of relapse, in terms of both longer-term costs and longer-term benefits to patients.

Overall, there was no substantial cost to the health-care system of discontinuing antidepressants; there were costs only to patients in terms of their well-being. Although the evidence from our analysis would not support recommendations at a national level to discontinue antidepressants based on cost-effectiveness, there is the opportunity to provide participants with the information that they require to make a more informed choice about their continued antidepressant prescription. There was a difference in utility scores at 3 months, although this difference had disappeared by 12 months, meaning that any detriment due to relapse or tapering the original medication, although significant, was, on average, short-lived. Some relapses, however, can be potentially severe and disabling, and our analysis offers no prediction of the impact on any particular person. Patients who wish to discontinue their antidepressant, and are willing to accept the risk of a shorter time to relapse, have the possible future benefit of eventually being medication free. The key costs would be mainly to themselves in terms of an increased likelihood of a relapse that could be potentially severe, for a short-term reduction in health-related quality of life, plus health service costs of an additional GP appointment to manage a potential relapse and other treatment options, such as cognitive–behavioural therapy. In societal terms, we also did not find a significant impact on productivity, partly because of the very wide CI around the result.

Strength and limitations

To our knowledge, the ANTLER trial is the largest individual randomised trial of antidepressant maintenance treatment that is not funded by the pharmaceutical industry. Clinical trials are often criticised for using narrow inclusion criteria, which can reduce external validity. 79 Our design ensured that we recruited people currently receiving long-term maintenance antidepressants in primary care. Participants had been receiving antidepressant medication for ≥ 9 months (the majority for > 3 years), which, to the best of our knowledge, was a longer treatment period than any previous trial. Our results are, therefore, more readily generalisable than the results of previous trials to the population currently receiving long-term maintenance antidepressants in primary care. We used the four most commonly prescribed antidepressants in primary care. Given that the vast majority of antidepressant prescriptions are for long-term treatment, including maintenance, we think that our results will apply to most people receiving maintenance treatment. We investigated three SSRIs, which have a similar pharmacological profile and act via similar mechanisms. Our results should, therefore, be generalisable to other SSRIs when used in this population. Although there are similarities in the mechanism of action of all commonly used antidepressants, it is more difficult to generalise our findings to other classes of antidepressant. We were unable to investigate whether or not treatment effects differed for mirtazapine because of the small number of participants receiving this antidepressant. We investigated the usual doses of antidepressants used in the UK. Our results should generalise to higher doses, given that there is no evidence of added efficacy,22 but it is difficult to generalise our findings to maintenance treatment at lower doses.

Attrition is a limitation of most randomised controlled trials (RCTs), and more participants in the discontinuation group than in the maintenance group dropped out. However, attrition was modest and results were unaltered after we adjusted for variables associated with missing data. Participants in the discontinuation group correctly guessed which group they were in more often than those in the maintenance group. This is consistent with prior studies suggesting that patients can distinguish placebo from active treatment. In principle, this could bias the outcome, but it could also be a consequence of relapse or withdrawal symptoms resulting from allocation to discontinuation (and on the causal pathway from exposure to outcome).

Although our design was embedded in clinical practice, it introduces some potential limitations. We had to ask retrospectively about past history of illness and treatment. In particular, we do not have detailed information of the original clinical decision for prescribing the antidepressant or any diagnostic information at that time. Participants who had more recently become well may have been more vulnerable to relapse, although this is unlikely to have biased the treatment effect owing to randomisation. Although our sample is likely to be more representative than prior trials, only a small proportion of those potentially eligible participated, which is common in RCTs. We also recruited people who had experienced at least two prior depressive episodes, so our findings do not necessarily apply to those receiving treatment for their first depressive episode. Participants were also mostly white, married and employed, and were recruited predominantly from moderately sized general practices in urban areas. Those who participated were older than those invited who did not participate. This limits generalisability but is still an improvement on existing literature and provides more realistic estimates of what might happen if someone on long-term maintenance treatment stops their antidepressant, at least for the first 12 months after discontinuation. A more important limitation is the ability to generalise to other health systems, although there are striking similarities in how antidepressants are used in wealthy countries. 80

In our trial, we used a novel assessment (the rCIS-R) to measure the reappearance of depressive symptoms. The results of our test–retest reliability study nested within the ANTLER trial provide strong evidence that the rCIS-R is a reliable measure of assessing reappearance of depressive symptoms. There was excellent agreement for definitions of relapse, the individual symptoms that were assessed and the sum of the symptoms scores. The main advantage of the rCIS-R is that, to our knowledge, it is the only fully structured interview assessing time to relapse. The pragmatic advantages that deserve considerations are time required for completion, simplicity of scoring and absence of any special training requirements.

Implications for health care

Our findings have several implications for the management of depression in primary care. Our evidence that discontinuing long-term maintenance antidepressants increases the risk of a clinically significant relapse in the following 12 months suggests that, for many patients, long-term treatment is appropriate. However, the severity of relapse varies. Although 56% of people who discontinued their treatment experienced a relapse, only around half of them (53%, 95% CI 44% to 62%) chose to return to an antidepressant prescribed by their clinician. It is possible that some relapses, or possible withdrawal symptoms, were not severe enough for the individual to decide that they needed to return to their medication. If six (95% CI 3 to 19) people stopped their medication, we estimated that one would experience a relapse who may not have experienced a relapse if they had remained on maintenance treatment. Our results illustrate that remaining on maintenance antidepressants does not guarantee well-being and is offset by any adverse events and the reluctance of many people to stay on medication for many years. If people who want to discontinue their antidepressants are regularly monitored in primary care during the first 6 months, relapse prevention may be possible with alternative treatments, such as psychological therapies. For example, there is good evidence that mindfulness-based cognitive therapy (with support to taper or discontinue antidepressants) is as effective as maintenance antidepressants at preventing relapse. 53

Current best practice is to engage with patients’ priorities and collaborate in coming to a decision about medication. For the individual patient, it is possible to know only about the average likelihood of relapse and that the severity of potential relapses will be unpredictable. Our findings will give patients and physicians an estimate of the likely benefits and harms of stopping long-term maintenance antidepressants to inform shared decision-making in primary care.

Recommendation for research

The ANTLER trial identified the following research needs:

• Further research into the longer-term implications of tapering antidepressant use compared with continuing with long-term maintenance for the prevention of relapse, in terms of both longer-term costs and longer-term benefits to patients.

• In addition to having longer follow-up duration, it would be beneficial if future studies were larger so that differences according to antidepressant type could be identified.

• Further research into distinguishing between withdrawal symptoms and depressive symptoms in the first few weeks of tapering antidepressant medication. This could involve experimental designs, new measurement instruments or factor analyses.

• The rCIS-R was designed for research purposes, although it may also have application in clinical practice as a simple way of assessing relapse in depression. This could be investigated in future studies.

Contributions of authors

Larisa Duffy (https://orcid.org/0000-0002-0093-3877) was the trial manager, conducted the analysis of the rCIS-R reliability data and was responsible for drafting and collating the report.

Caroline S Clarke (https://orcid.org/0000-0002-4676-1257) conducted the health economic analysis and with Rachael Hunter (https://orcid.org/0000-0002-7447-8934) drafted the health economic chapter.

Louise Marston (https://orcid.org/0000-0002-9973-1131) with input from Nick Freemantle (https://orcid.org/0000-0001-5807-5740) conducted both primary and secondary analysis and together with Gemma Lewis (https://orcid.org/0000-0001-6666-3681) carried out additional analyses and drafted the results section.

Gemma Lewis drafted the discussion section.

Simon Gilbody (https://orcid.org/0000-0002-8236-6983) had responsibility for the York site.

Tony Kendrick (https://orcid.org/0000-0003-1618-9381) and Michael Moore (https://orcid.org/0000-0002-5127-4509) had responsibility for the Southampton site.

David Kessler (https://orcid.org/0000-0001-5333-132X) and Nicola Wiles (https://orcid.org/0000-0002-5250-3553) had responsibility for the Bristol site.

Nick Freemantle, Simon Gilbody, Rachael Hunter, Tony Kendrick, David Kessler, Michael King (https://orcid.org/0000-0003-4715-7171), Paul Lanham (https://orcid.org/0000-0001-6864-3318), Michael Moore, Nicola Wiles, Irwin Nazareth (https://orcid.org/0000-0003-2146-9628) and Glyn Lewis (https://orcid.org/0000-0001-5205-8245) were responsible for the original proposal securing funding.

Glyn Lewis was chief investigator of the trial and had clinical responsibly for the RCT.

All co-applicants with input from Larisa Duffy, Dee Mangin (https://orcid.org/0000-0003-2149-9376), Faye Bacon (https://orcid.org/0000-0002-7566-7585), Molly Bird (https://orcid.org/0000-0001-5570-1419), Sally Brabyn (https://orcid.org/0000-0001-5381-003X), Alison Burns (https://orcid.org/0000-0002-2242-3499), Yvonne Donkor (https://orcid.org/0000-0003-2716-7743), Anna Hunt (https://orcid.org/0000-0002-0864-4113) and Jodi Pervin (https://orcid.org/0000-0003-2452-2391) designed the trial and developed detailed protocol.

All authors have provided substantial contribution to the conception of the ANTLER trial, interpretation of data and had input into drafting the report and/or revising it critically for important intellectual content. All authors have given final approval of the version to be published.

Publications

Lewis G, Marston L, Duffy L, Freemantle N, Gilbody S, Hunter R, et al. Maintenance or discontinuation of antidepressants in primary care. N Engl J Med 2021;385:1275–67.

Clarke CS, Duffy L, Lewis G, Freemantle N, Gilbody S, Kendrick T, et al. Cost-Utility Analysis of Discontinuing Antidepressants in England Primary Care Patients Compared with Long-Term Maintenance: The ANTLER Study [published online ahead of print November 8 2021]. Appl Health Econ Health Policy 2021.

Data-sharing statement

We are open to collaborations with other scientists as our ethics permission includes only use of the data by the research team. Please address your enquires to the corresponding author.

Disclaimers

This report presents independent research funded by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, NETSCC, the HTA programme or the Department of Health and Social Care. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the HTA programme or the Department of Health and Social Care.