Abstract
BACKGROUND: External implementation support (EIS) is a well-recognized feature of implementation science and practice, often under related terms such as technical assistance and implementation facilitation. Existing models of EIS have gaps related to addressing practice outcomes at both individual and organizational levels, connecting practice activities to intended outcomes, or grounding in well-established theories of behavior and organization change. Moreover, there have been calls to clarify the mechanisms of change through which EIS influences related outcomes. METHOD: In this article, we theorize about mechanisms of change within EIS. Our theorizing process aligns with the approach advocated by Kislov et al. We aim to consolidate prior EIS literature, combining related constructs from previous empirical and conceptual work while drawing on our extensive EIS experience to develop a higher-order, midrange theory of change. RESULTS: Our theory of change is empirically and practically informed, conceptually situated within an established grand theory of change, and guided by eight practice principles and social cognitive theory. The theory of change proposes 10 core practice components as mechanisms of change within EIS. When used according to underlying theory and principles, they are believed to contribute to favorable practice outcomes at individual, team, organizational, and system levels. The model offers flexibility by recognizing the need for sequential support processes and the demand to practice in dynamic and responsive ways. Case examples are presented to illustrate major themes and patterns of the model in action. CONCLUSIONS: The proposed model is intended to support prospective EIS studies by conceptualizing discernable practice components with hypothesized relationships to proximal and distal practice outcomes. The model can be behaviorally operationalized to compliment and extend competency-based approaches to implementation support practitioner (ISP) training and coaching. Over time, the model should be refined based on new empirical findings and contributions from ISPs across the field.