The MRC framework for developing complex interventions brings focus on the need to understand and explicate what the active components are within an intervention, with theory development and context considered core elements [10]. The ToC development approach described here presents how the identification and review of the published evidence base and the understanding of the contextual issues around pain and exercise for PWH have been successfully integrated into synthesising a stakeholder informed programme theory for the development of a complex rehabilitation intervention. The resultant programme theory model is visually coherent and with the mapping of behaviour change theory to it, the stakeholders consider the theory to be plausible, credible and testable.

Complex interventions involve a number of interacting components that may require new behaviours by those receiving the intervention, or those who deliver it [22]. They may also have outcomes that are intended, unintended and multiple, and have implementation chains that can be long and convoluted [30]. It is this potential multitude and interlinking of known and unknown variables that creates the concept of the ‘black box’ in complex interventions, not just ‘what’ and ‘where’, but ‘why’ and ‘how’ observed effects may be taking place [31]. Living with a rare disorder such as severe haemophilia brings with it a complex medical regime needed to manage it, widespread musculoskeletal consequences of joint haemarthroses and the lived experience and beliefs of each individual. Acknowledging these multiple factors and the need to understand the degree of interplay between them, confirms that an exercise-based telerehabilitation intervention for PWH living with chronic pain can be considered a complex intervention.

Theory is a set of interrelated concepts and definitions that explain or predict events by specifying relationships between variables [32]. A programme theory describes how an intervention is expected to lead to its effects and under what conditions, and should articulate the key components and how they interact, and the relationship between the contextual influences and the mechanisms of interaction [10]. Rather than a more linear logic model, the outcomes chain diagram in a ToC places more focus on the causality through which the order of the activities is linked, thereby clearly identifying outcomes critical to success [33]. A well-articulated programme theory may optimise practice and provide accountability and efficiency for chosen interventions [34]. Relational detail such as that contained within the ToC model, can provide in-depth delivery knowledge and explanations informing whether the intervention needs to be modified for scaling up or used in different locations [35]. The ToC map created here identified that a trusting therapeutic relationship was key to accepting that this approach may be helpful in the overall delivery of the rehabilitation intervention. The stakeholders were very aware of the different social and physical contexts of potential participants with haemophilia and arthropathy, and as a result agreed the intervention should focus on low impact, whole body movement and activity, rather than be joint/limb specific. In highlighting this activity approach as an enabler, it was felt to be most likely to achieve the medium-term outcomes in confidence, knowledge, participation, and enjoyment of the programme.

Stakeholder engagement is a core element of the MRC framework in complex intervention development. Involvement of stakeholders from the outset is vital, as they understand the problem at hand and can identify the priorities in order to find realistic, workable and meaningful solutions [22]. Such an approach may be particularly worthwhile in rare conditions such as haemophilia where stakeholder involvement can improve interventions design and meaning as well as reducing research waste. Whilst there is little evidence in the current literature pertaining to involvement of PWH for intervention theory development, there have been other successful examples of participatory approaches in developing methods to improve haemophilia care delivery. Timmer and colleagues worked with stakeholders (PWH and primary care physiotherapists) to explore their experiences of primary care and develop recommendations to optimise physiotherapy care co-ordination [36]. In approaching this problem this way, they were able to get consensus on 13 recommendations for better physiotherapy care that may improve service quality and reduce waste. Similarly, a pain treatment planning questionnaire was conceptualised with PWH and carers. The tool was developed in partnership with patients interviewed to guide and inform the content, which was then further refined after clinical testing using a ‘Think aloud’ approach. The authors noted that the co-design approach was instrumental in developing the condition specific checklist within the questionnaire that was also acceptable for the patient population it was tested on [37]. Given the scale of potential benefits when inclusive stakeholder approaches are used, it is unfortunate there is little current evidence of such approaches being used to develop rehabilitation interventions for PWH.

The integration of behaviour change approaches in physiotherapy research and practice has been identified as a necessity to develop future interventions related to health promotion and wellbeing [38]. For any intervention that proposes to change behaviour, the UK National Institute for Health and Care Excellence (NICE) recommends that the content of the intervention is specified, detail is provided about what is done, by whom and in what context and it is clear what underlying theory will be used to make explicit the key causal links between actions and outcomes [39]. Whilst no studies in haemophilia to date have used this approach of mapping COM-B to qualitative findings, it has been identified as having potential importance in designing meaningful exercise based interventions for those with living with multi-morbidity [40]. A clearly documented development process is important to add to a growing body of knowledge about the explicit methods used in developing interventions [41], benefiting an understanding of the implementation as well as being able to examine the generalisability of the techniques used [24].

The approach to stakeholder participation described here is novel in terms of previous approaches in PWH. With that in mind, it is important to include a reflective evaluation of their involvement in this process, and how their own interests and beliefs may have influenced and impacted the study.

All the PWH volunteers were white men, although they did have a wide age range and had a large diversity of experiences of haemophilia. One man had grown up in a country with minimal access to factor concentrate and comprehensive care, even when in adulthood. Another grew up with very intermittent and limited experiences of specialist physiotherapy. All of them had lived with pain associated with their haemophilia since childhood, but also had a positive view on the potential benefits of exercise. It was clear that all three men viewed their participation in this process as a philanthropic endeavour, viewing it as an opportunity to take part in something that might positively influence physiotherapy care provision for other PWH.

The two female physiotherapists who volunteered to be part of the stakeholder group were specialists from large treatment centres who had each worked in haemophilia for more than 10 years. Both had made contact with the lead author to volunteer their time for any projects associated with the study development. Similar to the PWH stakeholders, the physiotherapists did not expect a direct benefit from participating. They did note a desire to experience being part of an approach such as this, and an awareness that current approaches to pain management for PWH were insufficient.

The approach used in the theory of change workshop meant that all views were privileged as equal, enabling a safe and supported space for all suggestions to be talked through, and outcomes only reached by group discussion and group consensus. This focus kept the group coherent, and meant that the agreed suggestions had to be sensible and achievable in a real world setting, further adding to the impact of the theory of change model. The lead author moderated the workshop in the capacity as a critical friend, therefore sharing power and acknowledging all views to equally privileged.

This approach to stakeholder participation brought many benefits. It strengthened the focus on recruitment and delivery of the proposed telerehabilitation intervention, as well as highlighting which outcomes to evaluate within the study. The real world applicability of the proposed intervention, with a focus on a low impact/moderate intensity approach was probably the most impactful outcome of this process.

A major strength of this process is the transparency in the approach, with each step reflective of that which came before, and that which follows. The degree of detail regarding process means others can replicate it within their own environments, as well as being able to fully evaluate the process undertaken here.

The approach to co-production taken here for the ToC development is novel, but such an approach serves to shift the power dynamic away from the investigator and towards the stakeholders. The outcome of this has been a detailed, meaningful, and realistic theory that can be tested in real-world situations. The experiences and input of the stakeholders changed the intervention development for the better, and in doing so created a sense of ownership by them in that process.

Another strength is the clear, logical, informed process by which BCT’s were identified to be included in the study protocol. The BCT’s selection process can be situated in the synthesis and evaluation of the evidence base, as well as the mapping process onto the ToC co-produced by the stakeholders.

A limitation of this process may be the relatively small number of people involved in the ToC process. However, given the time and financial constraint associated with the development of a small feasibility study, the size of the stakeholder group was felt to be adequate by the research team. The ToC process itself was highly reflexive by virtue of the method and the review process of the map itself, thereby increasing transparency in the decisions made.

The stakeholders here in this study were adults currently receiving haemophilia care within the Healthcare system of the United Kingdom. Therefore, the output of the ToC may not be wholly transferable to paediatric populations, nor others in a different medical care system with differing geographical and contextual settings and experiences. However, the ToC method in itself is transferable for use in other haemophilia patient populations, and we encourage other researchers to consider its use when designing intervention studies.

Another limitation may be that the ToC map itself may be observed by others outside of the process to be lacking detail, or be thought to be missing outcome chains, activities, or enablers. This view is acceptable, but it must be remembered that the process described here is done in such way so as to be transparent and open to change. This iterative ability is what makes this approach advantageous for use in a feasibility study.

Whilst not a limitation, it should be highlighted that the lead author received specific training in the method, design and delivery of a Theory of Change workshop. Those interested in this method of stakeholder engagement should seek local providers of such training or expertise when thinking about this approach within their specific populations.