a pilot trial of Group-MCT for patients with depression and/or anxiety

Some parts of the sections that follow have been reproduced with permission from Wells et al.16 and McPhillips et al.14,15 These are Open Access articles distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The text below includes minor additions and formatting changes to the original text.

Work stream 1 overview

Anxiety and depression are common among CR patients. However, existing psychological interventions used in CR produce only modest reductions in emotional distress. An alternative therapy currently not used in CR, MCT, has shown promising results in improving anxiety and depression in mental health settings and in patients with physical illnesses. WS1 aimed to evaluate the acceptability and feasibility of delivering Group-MCT to CR patients experiencing anxiety and depression.

Work stream 1 was a multicentre pilot feasibility study with 4- and 12-month follow-up comparing Group-MCT plus usual CR (intervention) with usual CR alone (control). The study was designed as an internal pilot of the full-scale RCT (WS2). Prespecified criteria for progression to a full RCT were (1) a mean recruitment rate of 8.7 per month, with a rate of 10 per month being desirable; (2) ≥ 65% of participants in the MCT arm attending at least four of the six Group-MCT sessions; and (3) 75% retention at 4-month follow-up. Additionally, the pooling of data collected under the pilot with those collected under the full RCT in the final analysis depended on no substantial changes being made to the trial procedures (e.g. patient eligibility criteria, follow-up schedule) or the trial instruments (e.g. outcome measures) as a result of the pilot. Both study progression and the pooling of data sets required the agreement of the TSC and NIHR as the funding body. The study was approved by the National Research Ethics Service of the NHS (reference 15/NW/0163) and registered with a clinical trial database (ISRCTN reference ISRCTN74643496).

Collaboration with our patient and public involvement (PPI) advisory group took place throughout the study, with PPI members involved at every stage. WS1 recruited 52 CR patients who had elevated anxiety and depression scores on the HADS. Patients were recruited from three NHS trusts in north-west England: University Hospital of South Manchester NHS Foundation Trust, Central Manchester University Hospitals NHS Foundation Trust and East Cheshire NHS Foundation Trust.

The results of the pilot study provided evidence that Group-MCT was acceptable and feasible to deliver within CR. With the agreement of the TSC and NIHR, we progressed directly to a full-scale randomised trial of adding Group-MCT to CR (WS2). The pilot study found that no substantial changes to the trial procedures or instruments were required; therefore, agreement was also given to merge the pilot and main trial data in the RCT analysis. The re-estimation of the full trial sample size based on the pilot data and available resources resulted in a decision to increase the total recruitment target to 332, providing 90% power to detect the desired 0.4 effect size.

Aims and objectives

1. Confirm procedures for recruitment, randomisation, intervention delivery and data collection prior to a full-scale trial.
2. Collect data on recruitment and retention rates, and variability and clustering, in outcome measures, to confirm the sample size calculation and timeline for the full-scale trial.
3. Obtain preliminary economic data to inform economic modelling in the full-scale trial.
4. Include outcome data in analysis of the full-scale trial if no changes are required to the key features of the trial following the pilot.
5. Interview Group-MCT patients, including those who declined to participate or dropped out, to assess their (1) emotional experience since the index event, (2) interaction of emotional state with clinical care, (3) reactions and expectations on being offered the intervention and (4) for those engaged, their perceptions of the intervention.
6. Interview control patients to assess their (1) emotional experience since the index event and (2) interaction of emotional state with clinical care.

Methods

Work stream 1 was delivered in accordance with the grant proposal and employed a randomised pilot feasibility study with 4- and 12-month follow-up comparing Group-MCT plus usual CR (intervention) with usual CR alone (control). The study was designed as an internal pilot of the full-scale RCT (WS2).

Fifty-two CR patients with elevated anxiety and/or depression were recruited to a single-blind randomised feasibility trial between July 2015 and February 2016 from three NHS trusts in north-west England (University Hospital of South Manchester NHS Foundation Trust, Central Manchester University Hospitals NHS Foundation Trust and East Cheshire NHS Foundation Trust). The target sample size was originally 50 patients (25 per arm), determined as sufficient to evaluate recruitment and retention rates for a full-scale trial as well as rates of completion of the intervention. This sample was also adequate for estimating variability in outcome measures for which samples of 40 are generally considered sufficient. However, parallel recruitment across sites meant that 52 patients had consented by the end of the recruitment period and were included in the sample.

After giving informed consent, patients were randomly allocated to a trial condition in a 1 : 1 ratio using a minimisation algorithm that incorporated a random component in order to maximise balance between the arms in sex distribution, HADS anxiety and depression scores, and hospital site. Randomisation was conducted via a telephone link to the Manchester University Clinical Trial Unit (Manchester CTU).

The acceptability and feasibility of adding Group-MCT to CR was evaluated with respect to recruitment rates; attrition by the primary end point of 4 months; number of MCT and CR sessions attended; completion of follow-up questionnaires; and ability of the outcome measures to discriminate between patients. The study was also used to re-estimate the required sample size for a full-scale trial. We also examined the extent to which non-specialists in mental health (i.e. CR health providers) adhered to the Group-MCT protocol. For details of the data collection method, see the protocol.17

Trial population

The following inclusion criteria were applied:

1. Fulfilment of Department of Health and Social Care and/or BACPR CR eligibility criteria (acute coronary syndrome, revascularisation, stable heart failure, stable angina, implantation of cardioverter defibrillators/cardiac resynchronisation devices, heart valve repair/replacement, heart transplantation and ventricular assist devices, adult congenital heart disease, other atypical heart presentation)
2. A score of ≥ 8 on either the depression or the anxiety subscale of the HADS35
3. Age ≥ 18 years
4. A competent level of English-language skills (able to read, understand and complete questionnaires in English).

The following exclusion criteria were applied:

1. Cognitive impairment that precludes informed consent or ability to participate
2. Life expectancy of < 12 months
3. Acute suicidality
4. Active psychotic disorders
5. Current drug or alcohol abuse
6. Antidepressant or anxiolytic medications initiated in the previous 8 weeks
7. Concurrent psychological intervention for emotional distress.

Cardiac rehabilitation (treatment as usual)

Usual CR programmes comprise two components: exercise and educational sessions. CR programmes vary in content by site; however, all participating sites are offered core components (BACPR Standards and Core Components2) primarily using group-based delivery as part of outpatient provision in hospital or community settings supported by a multidisciplinary team. CR programmes across all sites ran weekly over 8–10 weeks. Exercise sessions were delivered in groups, with a therapist-to-patient ratio of 1 : 5 for low- and moderate-risk patients and 1 : 3 for high-risk patients. Educational seminars lasted 45–60 minutes and covered lifestyle and medical risk factor management. Additionally, all sites provided psychosocial intervention, including stress management and relaxation talks. Relaxation sessions at all sites included breathing techniques and progressive muscle relaxation. Two sites delivered psychoeducational talks on stress, while three sites included cognitive therapy methods for stress management (i.e. challenging negative thoughts, worry decision tree, behavioural activation). One site offered a 4-week stress management course as part of CR.

Group-metacognitive therapy

The MCT + CR intervention received group-based MCT in addition to the usual CR programme at their site. Group-MCT was delivered in six sessions lasting 60–90 minutes each, held once per week and facilitated by two CR professionals (i.e. physiotherapist, CR nurses and occupational therapists) or research nurses depending on the site. CR staff received basic training in implementing the treatment manual. Therapists completed a 2-day workshop delivered by the developer of MCT (AW). Training included didactic teaching, role-play, discussion and studying of the treatment manual. In addition, therapists delivered the intervention to a pilot group of volunteers along with an additional 1-day workshop that focused on enhancing initial skills. Therapists received ongoing supervision on an occasional basis while they were delivering the intervention.

Group-MCT focused on helping participants identify thoughts leading to the processes of worry, rumination and unhelpful coping behaviours. Participants were then guided through the practice of specific techniques to aid flexibility of and control over extended negative thinking patterns. Homework practice of the techniques was featured throughout the programme. At the end of treatment, patients received a ‘helpful behaviours’ prescription summarising what they had learned. Therapists’ adherence to the trial protocol was assessed through their completion at the end of each session of a checklist identifying the protocol components that had been implemented.

Data monitoring

Data monitoring, quality and handling were undertaken by the Manchester Clinical Trials unit and project oversight was conducted by an independent TSC.

Analysis

We assessed the feasibility and acceptability of adding Group-MCT to usual CR.

Feasibility outcomes included:

1. Completion of follow-up questionnaires (proportions of missing values, both overall and within-trial arms)
2. Ability of the outcome measures to discriminate between patients (range of scores, floor or ceiling effects)
3. Re-estimation of the required sample size based on the findings of this study (number of recruited patients required to detect an effect size of 0.4 on HADS total score at 80% power, controlling for baseline scores and allowing for attrition and clustering of patients within therapy groups)
4. Therapist adherence to study protocol

Acceptability outcomes included:

1. Study recruitment rate (number agreeing to participate out of those approached, and number recruited per month)
2. Withdrawal or drop-out by the primary end point of 4 months (attrition rate)
3. Numbers of MCT and CR sessions attended
4. Therapist adherence to study protocol

Results

Feasibility and acceptability of a trial of Group-MCT

The results of the feasibility study have been published.16 Participants were recruited between July 2015 and February 2016, and 38% of eligible patients were consented and randomised to the study, resulting in a recruitment rate of approximately 6.5 patients per month. Fifty-two participants (33 male and 19 female) were recruited, with 23 participants allocated to Group-MCT + CR and 29 allocated to CR alone. The mean age of participants was 58.67 years (standard deviation 9.47 years, range 38–79 years).

Retention at both 4- and 12-month follow-up was reasonable. At 4-month follow-up, 72.4% of patients in the control arm returned follow-up questionnaires; one (3.5%) participant withdrew from the study, six (20.7%) participants did not return the questionnaires and one (3.5%) questionnaire pack was lost in the post. The return rate of the intervention arm questionnaires was 69.6%; four (17.4%) participants formally withdrew from the study, two (8.8%) participants did not return the questionnaires and one (4.4%) questionnaire pack was lost in the post.

All questionnaires demonstrated a good range of observed scores, covering the majority of the possible score range, and with little in the way of floor or ceiling effects.

The trial did not negatively impact on attendance at usual CR, which was much the same in both trial arms. Participants attended a median of six sessions, with 58.6% of the CR-alone arm and 52.2% of the Group-MCT + CR attending at least six sessions of CR.

Among those allocated to the Group-MCT intervention, 56.5% of patients (n = 13) attended at least four of the six sessions, 21.7% (n = 5) attended one or two sessions and 21.7% (n = 5) did not attend any sessions. Therapist adherence to the MCT treatment protocol was monitored using an adherence checklist. Therapists were asked to indicate if specific components of the intervention had been completed at each session. Adherence was high at an average rating of 98.2% across all sites, with all sites deviating from the protocol only once.

Progression to a full randomised controlled trial

The decision to progress or not to a full RCT was made based on the data available at the time of submission of the study milestone report. At this point all patients had been recruited but only 18 had reached the 4-month follow-up point. The results at this point differed somewhat from those given above for the full study sample. The study’s overall recruitment rate of 6.5 patients per month reflected a slow start, followed by an average of nine patients recruited per month over the final 3 months. Seventeen patients (94%) had returned the 4-month follow-up questionnaire. Of 10 patients in the Group-MCT + CR arm, 6 (60%) had attended at least four treatment sessions: although slightly below the target of 65%, the small sample made this figure subject to large uncertainty. On the basis of these results and other evidence for acceptability and feasibility, including the qualitative work with patients and therapists, the NIHR as funder agreed that the research could progress to a full-scale RCT. To address the shortfall due to slow early recruitment, the study was expanded to include an additional two sites.

Sample size

Under assumptions of 25% attrition, a correlation of 0.5 between baseline and follow-up outcome scores, mean therapy group size of 5.75, and intracluster correlation coefficient (ICC) of 0.05, we originally estimated that a total recruitment sample of 230 patients for the main trial would provide 80% power to detect a treatment effect size of 0.4. As no substantial changes were made to the trial procedures or instruments following the pilot, with the consent of our TSC and NIHR as the funder the decision was taken to merge the pilot data with those of the main trial. Considering the updated parameters from WS1 [a 35% attrition rate, correlation of 0.5 (unchanged), mean group size of 3 and ICC of 0.05 (assumed)] and available resources, we revised the total recruitment target to 332 to give the full study 90% power to detect the desired 0.4 effect size.

Conclusion

The results suggested that a full-scale trial of Group-MCT within CR was feasible and acceptable to deliver.

Qualitative evaluations

Study participants (intervention and control) were interviewed in semi-structured qualitative interviews to explore the potential enablers of/barriers to recruitment/retention at several levels: patient (e.g. attitudes to emotional needs and support), intervention (e.g. comprehensibility) and service (e.g. practices or staff communications that contradict or support the intervention).

Some parts of these sections have been reproduced with permission from McPhillips et al.14 and McPhillips et al.15 These are Open Access articles distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The text below includes minor additions and formatting changes to the original text.

Ethics approval was obtained from NRES Committee Northwest REC (reference 15/NW/0163). All participants provided written informed consent prior to being interviewed. Qualitative interviews were audio-recorded, transcribed verbatim and pseudonymised. Analysis varied depending on the research question.

Analysis of the data to determine CR patients’ emotional distress and psychological needs followed a constant comparative approach that occurred in parallel with interviews to develop a thematic framework. The framework was further developed with subsequent interview transcripts.14 The analysis was evaluated based on its ‘catalytic’ and ‘theoretical’ validity, whereby findings had practical implications and connected with broader theory. The analysis was inductive in that they presented features of patients’ accounts based on emerging data/transcripts rather than on the significance for a priori theories. Theoretical frameworks were drawn on after the analysis was complete to consider the implications of the findings.14,36

Analysis of qualitative interviews also aimed to understand patient distress from the perspectives of CBT and MCT and comprised three stages: inductive analysis followed by a constant comparative approach, and, finally, reviewing transcripts combining deductive and inductive elements.15 The exploration of patients’ experience of MCT used thematic analysis. Using a systematic approach, codes were produced by grouping similar concepts together to identify key themes.

Psychological experiences and psychological needs of cardiac rehabilitation patients

We conducted a qualitative study using semi-structured interviews of 46 CR patients who had elevated symptoms of anxiety and/or depression.14 The study aims included:

1. Understanding how distressed cardiac patients describe their emotional needs
2. Understanding how CR patients described and understood their distress
3. Exploring patients’ thoughts about how well they thought their current CR and routine care addressed their psychological needs and their views on the role of formal psychological interventions

Patients often described their emotional experience since their cardiac event as negative, reporting how they felt low in mood and often engaged in worrying about and dwelling on a range of concerns, including ones that were unrelated to their health and predated their cardiac event.14 We did not find differences in accounts between men and women or between patients from different centres.

While patients were found to worry about and dwell on a range of topics, which is in line with findings of previous studies,36,37 they also described how they believed that worrying was uncontrollable and harmful, and they worried about worry (a process known as metaworry). The concerns CR patients have about worry (i.e. metaworry) have not been described previously; however, they are central to the metacognitive model38–40 and clarify how CR patients’ distress might be better addressed.

Patients described how they wanted to ‘get back to normal’ and stop worrying.14 They felt that they lacked a way to achieve this other than waiting for time to pass, and when they did seek support, they sought reassurance from staff and peers to check that they were responding ‘normally’.14 However, the effects of reassurance were generally transient and, consistent with previous findings, appeared to have little benefit for patients with cardiac symptoms.41,42 A new and potentially important finding was that, despite wanting reassurance, most patients were reluctant to talk about their worries in the context of CR unless they had been previously socialised into psychological interventions.1 Furthermore, despite being troubled by worry, most were dismissive of stress management and guided relaxation techniques offered in the context of existing CR. These techniques seemed superficial and difficult to apply in real life and needed more practice than CR provided. Patients also noted that they associated CR primarily with exercise classes and physical rehabilitation, which is in line with previous research, which may be a barrier to using CR as a setting to support mental health.43

Using MCT may overcome some of the barriers associated with current psychological approaches in CR, as summarised in Figure 2.

FIGURE 2 Barriers faced in current CR psychological treatment and solutions provided by MCT: a visual summary of discussion in our qualitative study.14 (Figure not reproduced here; see the original article.)

Assessing the ‘fit’ of the metacognitive model versus the cognitive–behavioural model in dealing with anxiety and depression in cardiac rehabilitation

The acceptability of the metacognitive model for CR patients was assessed both alone and comparatively with a model more frequently used in NHS services, CBT.15 Although CBT achieves moderate effects in CR for patients experiencing anxiety and depression,7 it may be that a psychological theory that does not focus on the content of patients’ negative thoughts (e.g. ‘I might have another heart attack’) but instead focuses on regulating worry and rumination may be better able to moderate the diverse range of concerns linked to anxiety and depression. By comparing patients’ perspectives via an interview guide designed to assess their concerns and causes of distress, it became apparent that worry and rumination exacerbated distress and that this perseverative negative thinking often began with a realistic negative thought, for example ‘I’ll never get back to full fitness, I’ll never live a full life in the same way’. Although it is evidently the case that CR patients will experience such thoughts, it is not always the case that they will continue to think those thoughts. CBT seeks to challenge such realistic thoughts, which is frequently unachievable. MCT would enable patients not to further engage with such thoughts, thereby limiting the time spent ruminating and having the positive effect of reducing the extent of negative thinking. Findings from the study15 illustrated that MCT may have a better fit with the experiences of CR patients; this overall conclusion was based on a sample of 49 patients who took part in a thematic interview.15 This group of patients reported a diverse range of worries but were reluctant to discuss them, offering the ideal opportunity for MCT to be used to overcome the distress of these patients as with this approach there is no need to discuss the content of worries. Conceptualising patients’ distress from the perspective of CBT involved applying many distinct categories to describe specific details of patients’ talk, particularly the diversity of their concerns and the multiple types of cognitive distortion. It also required distinction between realistic and unrealistic thoughts, which was difficult when thoughts were associated with the risk or consequences of cardiac events. From the perspective of MCT, a single category – perseverative negative thinking – was sufficient to understand all this talk, regardless of whether it indicated realistic or unrealistic thoughts, and could also be applied to some talk that did not seem relevant from a CBT perspective.