Mindfulness-Based Cognitive Therapy for Depressed IAPT Non-Responders

Major Depression represents a pressing challenge for health care. The disorder is not only highly prevalent but also shows many characteristics of a progressive disorder. If not treated sufficiently, it tends to become more recurrent and chronic over time.

In order to address this challenge, it is imperative to provide treatments that effectively reduce symptoms in those who are affected. In England and other parts of the United Kingdom, Increasing Access to Psychological Therapies (IAPT) services have been introduced with the aim of providing evidence-based psychological therapies in a timely manner. While the introduction of these services has successfully increased access to psychological therapies, outcome reports indicate that about 50% of the depressed patients who have completed high-intensity evidence-based psychological therapies within IAPT do not reach recovery and continue to show significant levels of symptoms. At the same time, progression to secondary care remains reserved for those with complex depression and high risk for suicidality.

Mindfulness-Based Cognitive Therapy (MBCT), an eight-week, group-based intervention that combines intensive training in mindfulness and elements from cognitive therapy for depression, may be particularly suited for addressing this gap. While originally developed, for the prevention of relapse in remitted patients with a history of recurrent depression, recent research has brought promising evidence that MBCT can have significant beneficial effects in patients with acute symptoms, and particularly in those who have not responded to previous interventions. However, evidence is currently not sufficient to warrant endorsement for use as a further-line treatment within the evidence-based IAPT pathway.

In order to justify and enable this use, research needs to provide evidence from a further definitive trial. Moreover, previous research investigating MBCT as a further-line treatment has been conducted exclusively in non-responders to antidepressant medication. In order to provide representative estimates of benefits for inclusion in IAPT, research will have to test clinical effectiveness following psychological therapy and demonstrate its cost-effectiveness within this particular pathway.

This study will compare MBCT to treatment-as-usual (TAU) in IAPT high-intensity treatment non-responders in a definitive clinical trial. We will test the immediate effects of the intervention on depressive symptomatology as well as whether effects on symptomatology can be sustained over a period of six months, thus taking into account the high risk of relapse in early stages following treatment. In addition to testing clinical effectiveness, the investigators will measure service use and collect information on quality of life in order to provide information on the cost effectiveness of the intervention in IAPT high-intensity treatment non-responders.

Design

The investigators will randomise 234 patients who have not responded to high-intensity IAPT interventions for depression, but do not meet eligibility criteria for secondary care services, in a 2-arm trial to take part in either MBCT or to continue with TAU, providing a comparator that is reflective of the current state of care (and in most cases will entail continued use of antidepressant medication). We will measure outcomes at baseline, 10-week and 34-week follow-up post-randomisation. Economic analyses will investigate effects of the interventions on subsequent service use.

Study setting

The research will be conducted at three main sites: South London and Maudsley NHS Trust, Sussex Partnership NHS Foundation Trust, and at the AccEPT Clinic, Mood Disorders Centre, University of Exeter, in collaboration with Devon Partnership NHS Trust.

Participants

The investigators will recruit depressed treatment non-responders to IAPT high-intensity treatments into the study. For detailed criteria see further below.

Intervention

Participants will be allocated to receive either MBCT or TAU. MBCT combines mindfulness training with elements from cognitive therapy. The investigators will follow the treatment manual with minor adaptations to address the fact that patients are suffering from current symptoms of depression following practice from our previous research. MBCT consists of eight weekly group-based sessions and participants are asked to engage in home practice of mindfulness meditation for about an hour per day. The intervention will be delivered by trained MBCT therapists together with an assistant to groups with a target size of 13 patients (minimum of 8 and maximum of 16) using videoconferencing on a secure online platform.

Participants in the TAU condition will be asked to continue with their usual care and follow the regimes suggested by their GP or mental health professional, which in most cases will consist of continuing use of antidepressant medication. Following previous practice in our trials, TAU participants will be invited to an interview to prevent tendencies towards 'resentful demoralisation' and highlight the importance of their contribution.

Economic evaluation

The economic evaluation will take a health and social care perspective, as required for evidence presented to NICE. In addition, the cost perspective will be broadened to include the costs of time off productivity losses, since these are known to be relevant and important in those attending IAPT services. Costs will be calculated by collecting service use information using the Adult Service Use Schedule (AD-SUS), as self-report measure developed by the team at King's and used in previous trials of MBCT, modified for use online, to which routine unit costs will be applied. The investigators will collect data on all service use not just use related to mental health conditions, because there is evidence that successful treatment in IAPT can reduce use of all healthcare services. Data on the use of the MBCT intervention will be collected via therapist records and costs estimated using the standard approach set out by Curtis et al., acknowledging the challenges of costing group-based interventions. Outcomes for the economic evaluation will be QALYs, calculated using health utilities derived from the EQ-5D-5L. Costs and outcomes will be combined first in a cost-utility analysis using QUALYS and second in a cost-effectiveness analysis using the PHQ-9, providing information on whether or not MBCT is worthwhile in terms of costs savings elsewhere or improvements in outcomes, and information will be provided to decision makers with statistical analysis of differences in costs, cost-effectiveness planes and cost-effectiveness acceptability curves.

Qualitative analyses

Qualitative analyses will be used to explore patient experience of the intervention and to understand how the treatment might best benefit patients in the IAPT pathway. The investigators will investigate:

1. patients' views on acceptability of MBCT and mindfulness practice,
2. patients' views of the changes they experience and their utilization of mindfulness skills, and
3. patients' views of the broader impact of MBCT on their lives.

A subsample of participants in the MBCT arm, estimated to be 24 (or until data saturation has been reached), will be invited to a qualitative telephone interview conducted by trained research assistants. Recruitment will be purposive, including patients across all sites, and seeking to achieve maximum variation in relation to:

1. completion/non-completion of treatment,
2. response/non-response to treatment, and
3. recruitment site (to examine contextual factors). Written feedback provided in the protocol sheets that MBCT participants receive on a weekly basis will be used to inform subsampling and will also provide us with the opportunity to explore any unanticipated experiences and effects in more depth.

Interviews will be audio-recorded, transcribed verbatim, and anonymized. Thematic analysis of interview transcripts will be conducted using a Framework approach, involving the coding and sorting of textual units according to both deductive and inductively-derived categories, and the use of matrices to review the coded data, investigate commonalities and differences and search for patterns.

Recruitment

Participating IAPT services will identify patients who are coming to the end of their high-intensity treatment and have not responded (PHQ-9≥10) or patients who within a 6-month window following the end of high-intensity therapy show levels of symptoms above caseness without prior remission.

Randomisation

The investigators will allocate individual participants to either MBCT or TAU at a ratio of 1:1 through remote randomisation at the UKCRC-registered Exeter Clinical Trials Unit (ExeCTU), following informed consent, completion of baseline assessment and enrolment in the trial. Randomisation will use minimisation on depression severity (PHQ-9<19 versus ≥19), antidepressant use at baseline and recruitment site.

Data collection methods

At baseline, trained research assistants will administer clinical interviews and self-report questionnaires at each of the sites. Post-treatment and follow-up assessments will be conducted remotely by asking participants to complete self-report questionnaires via secure online portal.

Blinding

As baseline assessment of participants is carried out prior to randomisation, there is no risk of disclosure of treatment allocation to the assessor at the time. Use of remote assessments, initiated through automated email, will rule out any potential effects of assessors on assessments of outcomes. The statistician analysing outcome data will remain blind to treatment allocation throughout the analysis.

Data management

Randomisation, data management, and quality assurance will be undertaken by ExeCTU under the supervision of the principal investigator, trial statistician and quality assurance manager.

Statistical methods

All analyses will be carried out using an a priori statical analysis plan as agreed with the TMG and TSC.

Participant characteristics at baseline (including number of previous depressive episodes and IAPT service) will be set out descriptively by treatment arm. The primary analysis approach will use the intention-to-treat (ITT) principle (all participants will be included in the analysis according to their randomised allocation irrespective of the treatment actually received) including observed data only. All outcomes will be reported descriptively at baseline, and at 10 and 34 weeks' follow-up. Continuous outcomes will be analysed using linear regression models. The binary outcome variables will be analysed using logistic regression. All analyses will adjust for participant covariates used in randomisation, with adjustment for baseline scores for continuous outcomes. We will assess other participant characteristics at baseline and will consider adjusting for any covariates that are found to be substantively unbalanced, should such covariates be considered predictive of outcome. Inferential between group comparisons (MBCT vs TAU) for the primary and all secondary outcomes will be performed at 34-week follow-up. As a sensitivity analysis, the investigators will perform a complier average causal effect (CACE) analysis for continuous outcomes only, to estimate the treatment effect while accounting for non-adherence to treatment. Mixed effects regression models with a random effect on individual participant will be performed for continuous and binary outcomes, including participants with outcome data reported for at least one follow-up time. Sensitivity analyses are described in the protocol and statistical analysis plan.