Why we adapt: an integrative qualitative study of context and intervention adaptations in facilitated implementations in primary care

Abstract Background Understanding and managing intervention adaptations during implementation is critical to maintaining fidelity to evidence while improving intervention fit in context. Implementation science faces a challenge in creating pragmatic methods to track and guide adaptations during implementations. In this study we demonstrate a pragmatic intervention documentation and tracking tool: the intervention menu. This tool allowed practice facilitators to track adaptations throughout implementation, and we collected and analyzed additional qualitative meeting data to characterize relationships between adaptations and context. Methods We conducted a qualitative descriptive study embedded within an EvidenceNOW cluster randomized hybrid implementation-effectiveness trial (NCT05664451). Data were collected from facilitated implementation meetings with clinicians and staff from 15 primary care clinics. Practice facilitators completed intervention menus after each meeting to document interventions and adaptations over time. Analysis proceeded in three stages. First, we identified intervention adaptations by comparing intervention menu entries across implementation timepoints. Second, guided by the Consolidated Framework for Implementation Research (CFIR) 2.0, we used deductive content analysis of mid-implementation meeting transcripts to identify contextual determinants relevant to adaptations. Third, we used inductive code mapping to integrate adaptation and contextual determinant data and characterize relationships between context and intervention adaptations. Results We identified two types of contextual influences on adaptations: (1) context drives adaptations and (2) context shapes the process of adaptation. The first captures how contextual factors (e.g., clinical workflows, patient population, information technology limitations) initially motivated, necessitated, or blocked specific adaptations. The second type captured how context continuously shaped the process of adapting interventions, influencing iterative refinements and maintenance of adaptations. Further, we demonstrated the utility of intervention menus, capturing 52 intervention adaptations total across 15 participating clinics. Conclusions This study identified two distinct types of relationships between adaptations and contextual determinants of implementation, using intervention menus as a pragmatic tool for intervention adaptation tracking. The themes identified in this study capture relationships between context and adaptations, going beyond initial drivers of adaptations to characterize how context continually influences adaptation processes. This work underscores the value of integrating diverse data sources to understand adaptations during implementation. Future studies will continue to evaluate the dynamic and complex relationships between context and adaptations.