Step One – Review of the literature
Dr Karen Heslop Marshall, a clinical academic agreed to be part of the research collaboration, allowing use of her CBT based Lung Manual (42), which had been evaluated in a successful randomized trial (39), as a basis for the development of the TANDEM intervention. We also recognised the importance of providing high quality, standardised information and education. To enable this, whilst allowing for the majority of our development time to focus on the CBA elements, Professor Sally Singh who had developed a self-management programme the SPACE manual for COPD (43) was invited to join the research collaboration. We therefore used SPACE materials as well as information leaflets from the British Lung Foundation to meet educational and self-management needs.
Step Two – Identification of Theoretical Underpinnings and Mapping to the intervention
Both The Lung Manual and the SPACE manual are evidence-based and draw on theory. The Lung Manual applied Beck’s theory of CBT (44) for managing anxiety and breathlessness in COPD whilst the SPACE manual applied a self-management approach based on Bandura’s social learning/cognitive theory (45, 46). Review of other theoretical constructs predictive of outcomes in COPD suggested that Leventhal’s Self-Regulation theory (47) was also relevant. Figure two shows the logic model for TANDEM. The basic premise is that how an individual thinks about their COPD (cognitions - including illness and treatment beliefs) influences how they behave (including self-management actions taken) and how they feel (both physical symptoms and emotions). These factors have interactional effects such that depression and /or anxiety can be both reduced or exacerbated depending on the individual’s cognitions and behaviours. Consequently, by targeting change at a cognitive, behavioural or symptom level this will influence emotional outcomes.
Consultation with Education for Health, a health education charity, and a review of the literature suggested that the pedagogical theory that would be relevant for the training of facilitators was the VARK (Visual, Auditory, Read, Kinesthetic (i.e. experience or practice, simulated or real)) model of learning (48). This ensured that the training would suit individuals with different learning styles.
For intervention development the person-based approach (49) with its focus on using qualitative work to inform guiding principles was considered particularly relevant and therefore guided the intervention development process.
Step Three – Qualitative findings, themes and development of guiding principles
One focus group comprising six rHCPs (one respiratory consultant, one occupational therapist, three physiotherapists and one exercise practitioner) and seven individual interviews (four psychologists, two physiotherapists and one general practitioner) were conducted. All had experience in working with patients with COPD, either in the community or secondary care. Roles varied, including some with management responsibilities who were able to discuss implementation.
One focus group was held with patients of whom four had COPD, two had other respiratory conditions and two were carers. Very tight timelines precluded formal analysis of transcripts from the patient focus group data, so limiting data to quotes selected from the audio recording. Major local governance delays prevented conduct of the second focus group with patients and their carers who had experience of CBT in time to inform intervention development.
Both patients and professionals presented an overall positive attitude to the idea of the TANDEM intervention:-
And I do think in the long run something like this could be more cost effective and stuff, things like that…I think it would be really useful (rHCP FG002 Occupational therapist)
Themes were developed which related to i) life with COPD, ii) intervention considerations and iii) issues for implementation (see supplementary article 1 for qualitative data summary).
Life with COPD
All participants including patients and rHCPs recognised depression and anxiety as common in COPD although patients did not always use this terminology. Other issues such as frustration and embarrassment, along with role adjustment and loss, including of social contacts, were common and seen as contributors to mood problems.
“The approach is good…because of all the other thing happening in people’s lives which can stop them attending PR and make them anxious and depressed.” (focus group patient)
rHCPs emphasised the challenge of breathlessness to patients with COPD and suggested that discussion and teaching of breathing control early in the intervention is a helpful way to raise issues around mood as well as providing practical help which may increase engagement. There was also recognition that this group may be quite socially isolated and health literacy may vary so the intervention must be accessible to all.
The majority of issues for implementation related to the workforce and who would realistically be able to deliver a CBA service. Both rHCPs and psychologists recommended rHCPs to be best placed, however all recognised that there would need to be some selection process and supervision. One clinician was concerned about the cost of the intervention, although others considered that in the long run CBA could be cost-effective.
These themes were subsequently interpreted to provide guiding principles as recommended by the patient based approach to intervention development (49). Table 1 shows these principles and example data extracts.
[Table one about here]
Step Four: Design and refinement of intervention
Having agreed the guiding principles for the intervention the intervention working group met to discuss the design of the TANDEM intervention. It was understood that the intervention would be working at two levels i) patient-facing (i.e. CBA delivery) and ii) facilitator-facing (i.e. training programme).
For the patient-facing CBA a range of materials were designed that could
be provided as part of the intervention. These were developed for TANDEM but where possible drew on or used the SPACE manual (43)
handouts. LS developed all materials with iterative refinement from the working group and PPI consultation. Each patient was given a TANDEM folder, in which they could store handouts relevant to them so that individuals had a tailored version of TANDEM materials whilst maintaining consistency in the content provided. In total 6-8 face to face sessions were designed, covering nine topics; those delivered were dependent on individual problems and complexity. Table two provides an overview of TANDEM topics
[Table two about here]
One topic (dealing with ‘other problems’) was specifically added as a strategy for keeping the focus of initial sessions on COPD whilst having space later to address issues the person may have outside COPD, for example debt, substance abuse etc. This topic looked at how the CBA, learnt in the context of COPD, could be generalised to different problems, with sign-posting to additional sources of help. At the final session discussion was around pulmonary rehabilitation. If there was to be a delay in individuals commencing PR then the facilitator arranged up to nine weekly telephone calls.
The three day facilitator training was provided with a supporting manual which covered the skills needed to deliver TANDEM.
[Table three about here]
Table 3 details the content of the programme. There was a high level of practical and experiential learning in the group and supportive links within the group were encouraged. Throughout delivery of the CBA intervention facilitators received telephone supervision, one-to-one with senior cognitive behavioural therapists at approximately fortnightly intervals. This on-going supervision was considered to be an integral part of the intervention.
Phase Five – Real World Pre-Pilot study
All three TANDEM facilitators completed the three training days and two went on to deliver the intervention to three patients (one delivered it to two patients and one to a single patient). The third facilitator (respiratory practice nurse without prior CBT training) did not manage to see any patients due to delays because of research governance and consequent change in work commitments.
After the initial training session, facilitators suggested some changes, specifically an overview of the intervention at the beginning of the training in order to orientate individuals. All the facilitators felt that a role-play activity with a simulated patient (actor), which was conducted as part of the original group training on Day one, was too threatening and at too early a stage of skill development. Instead they requested more demonstrations and more practice with developing formulation.
Interviews at the end of delivering TANDEM to patients, revealed that both facilitators felt the intervention had been well received by participants and feasible to implement, although one had to deliver it over a longer period than scheduled due to patient illness.
‘Yeah, I mean the two patients who I had were very, very enthusiastic about all elements of the intervention. (PPHCP01)’
Generally the facilitators appeared able to follow the manual and found it a helpful guide, but there was questioning of whether someone without previous CBA training would be able to manage:
‘I mean section nine, it's got identifying maintenance factors, and it talks about safety behaviours, avoidance and escape, catastrophic interpretation, scanning or hypervigilance, self-fulfilling prophecies, fear of fear, reductions, affectionism, short term rewards. If you're trying to talk to a patient and remember what it says in the manual you might get yourself a little bit flustered.’ (PPHCP02)
One facilitator recommended presenting basic intervention techniques as a toolbox and also the provision of a crib sheet for easy prompting within sessions.
‘I feel that people who come away from the training need to have something like a virtual toolbox of techniques that they can refer to …they expected quite a lot of you… I made myself a crib sheet type of thing’ (PPHCP01)
One element that was not adhered to as planned was supervision with a senior psychologist, as the facilitators relied on supervision by an experienced member of their team who was already known to them and who was also part of the study team (KHM). However both facilitators reported this supervision was useful.
Patients who had received the TANDEM intervention reported it to be acceptable and beneficial, observing that the facilitators had very good interpersonal skills. There were no substantive suggestions for improvement.
‘And then x and I just seemed to get on very well, he's a likeable chap, very laid back. And so it went from there. And then we started doing the things that you asked in TANDEM. Planning… They're just small things, but marvelous’ (PPP01, male participant)
Refinements to TANDEM after the pre-pilot
Refinements to the TANDEM intervention
Changes and additions were primarily made to the TANDEM facilitator training, as there were few recommendations for changes to patient materials. These are detailed in Table 3 with additions highlighted in bold. All suggestions were followed: e.g. providing a greater overview of TANDEM at the beginning of session one, keeping to a core set of CBA techniques and outlining a “toolbox” of techniques which could be used. We made video recordings to demonstrate therapeutic skills and CBA techniques. These were made available online, with a facilitator chat facility for ongoing support.
The use of a simulated patient was omitted from the first two days of training and replaced by partnered role-play. The simulated patient role-play was, however, added to the end of day three. Each TANDEM facilitator was individually video-recorded conducting a cognitive behavioural assessment and feed back with the actor. Each video was subsequently assessed by LS and an independent assessor to ensure a minimum standard of competency (see study protocol(37)) was acquired. To enhance learning, and boost confidence, facilitators received one-to-one feedback on their video. A training session on the importance of supervision was added with reflection that supervision is a standard part of psychological training and practice (in contrast to more managerial supervision with which HCPs may be more familiar).
Refinements to improve implementation
To improve delivery of the intervention within the trial and future implementation within routine healthcare contexts, five features were added:
- Facilitators were provided with crib cards for use as prompts within sessions.
- An optional session was added for use when a break in sessions had become necessary (e.g. due to patient illness). The aim of this additional session was to refresh topics that had been covered before the break and re-establish current priorities.
- Some flexibility in the order of delivery of sessions was allowed reflecting the reality that some patients commenced PR before the end of the TANDEM sessions. It was stipulated that Topics 1-5(or 6) must have been conducted but that if necessary the final topic on expectations of PR could be brought forward as there was no sense in delivering this once PR had started.
- A structure for screening potential facilitators including a formal application with a curriculum vitae and telephone interview with one of the principal investigators was developed. The aim was to ensure only fully committed individuals who could meet the study-specific requirements (e.g. flexibility to travel, willingness to complete research modules and good clinical practice training) received training.
- A booster training session was designed to be delivered to facilitators if there were delays of 3 months or more between initial training and delivery of TANDEM.
For a description of the intervention following TIDieR guidance please see additional file 1.