Table 1 Guide for developing complex interventions – key steps and considerations

1. Plan the development process

1. Identify the problem to be targeted and refine understanding of it throughout the process (iterative as literature emerges).

2. Assess whether the problem is a priority.

3. Consider which aspects of the problem are amenable to change.

4. Ask whether a new intervention is really needed and if the potential benefit of the new intervention. justifies the cost of development

5. Determine the time needed to undertake intervention development.

6. Obtain sufficient resources/funding for the intervention development study.

7. Draw on one or more of the many published intervention development approaches, recognising that there is no evidence about which approach is best and apply flexibly depending on the problem and context.

8. Involve stakeholders during the planning process (see next Action).

9. Produce a protocol detailing the processes to be undertaken to develop the intervention.

1. The problem was defined as: Girls disengage and drop out of sports due to body image concerns at a disproportionate rate to boys.

2. Assessed the priority of the problem via:

       a). Scientific literature

        b). News and social media;

        c). Government policy.

3. Reviewed existing and monitored new literature into girls’ body image experiences in sport contexts.

4. Body image is a malleable trait that can be improved through concerted intervention. Therefore, it was proposed that a body image intervention that is specific to sport contexts would enhance girls’ body image and sports enjoyment; thus, reducing the gender disparity in sports participation.

5. Our systematic review and meta-analysis found that existing interventions that address the intersection of girls’ body image and movement experiences, including sport, were marginally effective at improving body image and ineffective at improving movement-related variables. Key limitations included:

       a). A majority of interventions used a risk reduction framework (i.e., reduce body dissatisfaction), rather than addressing both risk and protective factors (i.e., reduce risk, while enhancing one’s relationship with their bodies beyond satisfaction).

       b). No interventions incorporated CBPR into the development and/or selection of an intervention; thus, implicating intervention feasibility, including acceptability, implementation, integration, and efficacy.

6. A detailed research proposal, including project aims, deliverables, protocols (e.g., where and how to involve stakeholders), and timelines were presented to industry partners for review and approval.

7. The proposal was approved and subsequently funding was obtained via two industry partners.

2. Involve stakeholders including those who will deliver, use and benefits from the intervention

1. Work closely with relevant stakeholders throughout the development process: patients, the public, the target population, service providers, those who pay for health and social services or interventions, policymakers, and intervention design specialists.

2. Develop a plan at the start of the process to integrate public and patient involvement into the intervention development process.

3. Identify the best ways of working with each type of stakeholder, from consultation through to coproduction, acknowledging that different ways may be relevant for different stakeholders at different times.

4. Use creative activities within team meetings to work with stakeholders to understand the problem and generate ideas for the intervention.

1. A global multi-stakeholder partnership was formed between academics, community, and industry partners. The partnership provided access to the target audience and community (i.e., girls, coaches, sport organisations).

2. Stakeholder input was strategically approached during the research planning phase; thus, ensuring that appropriate stakeholders were approached and that their time and expertise were used efficiently.

3. Creative and efficient ways of working were identified for the respective stakeholders (e.g., weekly meetings between core partners; focus groups with girls and coaches).

3. Bring together a team and establish decision making processes

1. Include within the development team individuals with relevant expertise: in the problem to be addressed by the intervention including those with personal experience of the problem, in behaviour change when the intervention aims to change behaviour, in maximising engagement of stakeholders and with a strong track record in designing complex interventions.

2. It may be hard to make final decisions about the content, format, and delivery of the intervention, so only some team members may do this. There is no consensus about the size or constituency of the team that makes these final decisions, but it is important early on to agree a process for making decisions within the team.

1. A project team was acquired that had relevant expertise to ensure intervention feasibility, including acceptability, implementation, integration, and effectiveness.

2. Review periods were incorporated throughout the research timeline, which allowed for key stakeholders to review the intervention content, format, and structure.

4. Review published research evidence

1. Review published research evidence before starting to develop the intervention and throughout the development process for example, to identify existing interventions, to understand the evidence base for each proposed substantive intervention component.

2. Look for, and consider, evidence that the proposed intervention may not work in the way intended.

1. Prior to intervention development, a systematic review and meta-analysis was conducted to identify existing interventions that address the intersection of body image and sport among girls, and key intervention, participant, and methodological features that impact intervention effectiveness. Findings from this research further refined the proposed intervention.

5. Draw on existing theories

1. Identify an existing theory or framework of theories to inform the intervention at the start of the process, for example, behaviour change or implementation theory.

2. Where relevant, draw on more than one existing theory or framework of theories for example, both psychological and organisational theories.

1. Theories pertaining to positive and negative body image and embodiment informed the theoretical framework of the new intervention

2. Following the systematic review and meta-analysis and focus groups with girls, an intervention framework was developed which outlined the intervention themes and underpinning theories, and the format and structure. The theoretical perspectives and frameworks underpinning the intervention activities, included:

a). The Tripartite Influence Model for Body Image [24];

b). Positive Body Image and Embodiment Theories [13,14,15, 25, 26], including Agency and Functionality, Body Connection and Comfort, Body Appreciation, Body Functionality, Resisting Objectification, and Attuned Exercise;

c). Cognitive Dissonance [27].

6. Articulate programme theory Note. A programme theory describes how a specific intervention is expected to lead to its effects and under what conditions

1. Develop a programme theory. The programme theory may draw on existing theories. Aspects of the programme theory can be represented by a logic model or set of models.

2. Test and refine the programme theory throughout the development process.

1. Programme Theory: If given the opportunity to play sports, girls face unique body image-related challenges that impede their participation and enjoyment, relative to boys. Their sport experience is hindered, if not halted, by the promotion of harmful gender stereotypes, unrealistic appearance and athletic ideals, the sexual objectification of female athletes’ bodies, and the vilification of girls’ physical competence. This relentless evaluation of female bodies can lead to appearance preoccupation, surveillance and dissatisfaction among girls, which in turn can diminish their concentration, performance and enjoyment of sports. Lastly, these negative sport experiences are further compounded by coaches and sports communities being ill-equipped to deal with these matters. Body Confident Athletes will educate girls about and how to challenge gender stereotypes, and unrealistic appearance and athletic ideals within sport contexts (e.g., body talk). It will also upskill them with the knowledge and strategies to develop a deeper acceptance, appreciation, and respect for their body, and what it can experience when playing sports, beyond what it looks like. That is, girls’ attention will move away from “What does my body look like during sport?” and towards “What can my body do and experience”, which in turn will increase girls’ body connection and comfort.

2. The programme theory has been refined throughout the development process, including the consideration and incorporation of new literature, and review sessions with girls, coaches, community, and industry partners.

7. Undertake primary data collection

1. Use a wide range of research methods throughout, for example, qualitative research to understand the context in which the intervention will operate, quantitative methods to measure change in intermediate outcomes.

1. A wide range of research methods were utilised and will continue to be utilised throughout this research, including:

a. Systematic reviews and meta-analyses;

b. Qualitative assessment tools (e.g., semi-structured focus groups);

c. Quantitative assessment tools (e.g., validated measures to assess change in primary and secondary outcomes in the pilot and main trials);

d. Randomised controlled trials, including an initial pilot (e.g., assessing initial feasibility) followed by a main trial (e.g., assessing efficacy).

8. Understand the context

1. Understand the context in which the intervention will be implemented. Context may include population and individuals; physical location or geographical setting; social, economic, cultural, and political influences and factors affecting implementation, for example, organisation, funding, and policy.

1. Both formal and informal information gathering steps were undertaken to better understand the context under which the intervention would be implemented and integrated into sport contexts. This included:

a. Project scoping and planning meetings with topic experts from the core partnership group (i.e., academics, community, and industry partners);

b. Systematic review and meta-analysis of existing interventions, and identifying optimum intervention features (e.g., number and length of sessions);

c. Semi-structured focus groups with girls and coaches from six countries, and qualitative surveys with young women from an additional seven countries. Participants provided insights into country and cultural nuances needing consideration in the new intervention, including gender norms (e.g., girls are not expected to play sport), socioeconomics (i.e., access to resources; facility standards), religiosity (e.g., uniform accommodation). Where appropriate, this information has been considered and incorporated into the intervention. Alternatively, these considerations will be signposted and explained in a localisation toolkit, which will enable those implementing the intervention to adapt the content/delivery accordingly, while retaining the evidence-based mechanisms that make the intervention effective.

9. Pay attention to future implementation of the intervention in the real world

1. From the start, understand facilitators and barriers to reaching the relevant population, future use of the intervention, ‘scale up’, and sustainability in real world contexts.

1. From the outset, the research designs for both the pilot and main randomised controlled trials have been considered and refined, as these will reflect real-world dissemination, implementation, and integration. For example, the delivery of Body Confident Athletes is task shifted to coaches, as they were identified as both influential community members (e.g., have close relationships with girls) and can permeate their knowledge and skills to multiple teams and system levels. To ensure sustainability was maintained as a key objective, the RE-AIM model for evaluating interventions was employed, such that the intervention was assessed on a number of metrics beyond the standard ‘efficacy’ (i.e., acceptability, integration etc.).

10. Design and refine the intervention

1. Generate ideas about content, format, and delivery with stakeholders

2. Once an early version or prototype of the intervention is available, refine or optimise it using a series of iterations. Each iteration includes an assessment of how acceptable, feasible, and engaging the intervention is, including potential harms and unintended consequences, resulting in refinements to the intervention. Repeat the process until uncertainties are resolved

3. Check that the proposed mechanisms of action are supported by early testing.

1. Intervention development was an iterative process conducted between academic experts in body image and sport and key stakeholders, including girls, coaches, community, and industry partners:

a. Initially, an intervention framework was developed, which outlined the intervention theory and aims, key intervention features (e.g., number and length of sessions; activities), and the primary and secondary outcomes;

b. Next, the framework was reviewed and signed off by the academic, community, and industry partners;

c. Next, the framework was expanded upon, with an initial intervention prototype developed (e.g., a fully formed session guide for coaches; an athletes’ workbook);

d. Next, the intervention prototype was reviewed and signed off by the academic, community, and industry partners, and then by a small group of girls and coaches;

e. Next, minor modifications were made to the intervention (e.g., changing terminology for greater comprehension);

f. Next, the intervention was pilot tested in a small-scale cluster randomised controlled (pilot) trial, where further data was gained on intervention feasibility, including girls and coaches’ acceptability, intervention implementation, and integration;

g. To finish, the intervention will be tested in large-scale cluster randomised controlled trial, where similar data will be captured to confirm acceptability, implementation, and integration, as well as efficacy

11. End the development phases

1. There are no established criteria for stopping the intensive development phase and moving on to the feasibility/pilot or evaluation phases. The concepts of data saturation and information power may be useful when assessment of later iterations of the intervention produces few changes.

2. Describe the intervention to facilitate transferability of an intervention outside the original team and location in which it was developed.

3. Write up the intervention development process so that judgements can be made about the quality of the process, links can be made in the future between intervention development processes, and the subsequent success of interventions, and others can learn how it can be done.

1. The large-scale cluster randomised controlled trial will provide additional data for the intervention feasibility, including acceptability, implementation, and integration. This data will be used to further refine the intervention.

2. A comprehensive overview of the intervention development process is provided in this study protocol, which will be available via an open-access journal. The intervention materials will also be freely available to the public, and therefore will be available for review and potentially additional evaluations beyond the current research team.