Planned adaptation and implementation of the

Background: The purpose of this paper is to report on the implementation of an evidence-based model, VIVA, which was developed to translate physical activity (PA) recommendations to rural environments and was scaled-up to 12 rural communities across New Mexico. Our longitudinal qualitative research describes processes of planned adaptation in the rural context with an exploration of inner and outer context adaptations that consider important implementation constructs including leadership, partnership and collaboration. Materials & methods: An enhanced version of the RE-AIM framework was used to formulate community-level engagement and process questions essential to implementation science. Qualitative methods, using a thematic approach that included both inductive and deductive coding with attention to processes, was used to explore adaptation at the community level. Data included semi-structured interviews with 17 community leaders at baseline and 10 at follow-up, fieldnotes, and technical assistance tracking forms. Analysis was conducted with NVivo qualitative data analysis software. Results: Analysis demonstrated how planned adaptation of the implementation model was critical to dissemination in rural communities. Understanding and adapting to local context-including geography, culture, economics-is essential for implementation. Inner context constructs, recognized as important across implementation models, including leadership, partnerships and political engagement were found to be key to implementation success. Moreover, we provide concrete examples of the range and complexity of these issues in rural communities, and how these shaped implementation uptake and success. Discussion: Studying processes of planned adaptation in rural contexts will further implementation science efforts to move evidence into practice. It is essential to incorporate planned adaptation to local, community contexts to create models which are simple to encourage adoption, are evidence-based, and are adaptable to local conditions without compromising the integrity of the evidence-based model.