We gained rich data on local context from the stakeholder FGs, particularly relating to the cultural and religious practices of the communities within the study population, which shaped the intervention design. The importance of understanding the cultural and religious

context in minority ethnic communities has been highlighted in other studies. In a childhood obesity prevention study targeting minority ethnic communities in London, Depsipeptide Rawlins reported child and parent perceptions of healthy eating and physical activity. The findings relating to South Asian communities resonate strongly with our data, for example the influence of places of worship and the role of extended family members on healthy lifestyles (Rawlins et al., 2013). A recent comprehensive evidence synthesis review on adapting health promotion programmes (including diet and physical

activity) for minority ethnic groups also draws attention to the importance of tailoring to particular contexts. The authors concluded that such adaptation Decitabine purchase increased intervention relevance and acceptability, although whether this results in increased effectiveness is undetermined (Liu et al., 2012). The need for considering local context brings up the question of intervention transferability to different settings. Hawe and colleagues argue that a complex intervention can be standardised and transferable if it is the function and process of the intervention (e.g. mechanisms to increase children’s physical activity in school) that are standardised rather than the components (e.g. a specific curricular activity). This enables the delivery of interventions to take into account

local context (Hawe et al., 2004). This approach necessitates a theoretical understanding of the change mechanisms of local context at each intervention site. We would argue that this is a viable approach. An understanding these of the contextual factors is essential for tailoring intervention components and thus determining their success. For example, barriers to childhood obesity prevention interventions, such as lack of parental time repeatedly emerge in the literature (O’Dea, 2003, Pocock et al., 2010, Power et al., 2010 and Sonneville et al., 2009). However, this barrier can only be addressed if the precise nature of the constraints on parental time is understood. In this study mothers were likely to be constrained through obligations such as looking after extended families or attendance at places of worship (Pallan et al., 2012), whereas in a North American study of white middle class children, perceived time constraints related to parents’ work commitments (Power et al., 2010). Different approaches to intervention would be required to overcome this barrier in these two communities. The iterative development process enabled us to implicitly gain a theoretical understanding of change pathways, and use this to drive intervention development.