Translating evidence in complex systems: a comparative review of implementation and improvement frameworks

Abstract

Purpose An increasing number of implementation and improvement frameworks seek to describe and explain how change is made in healthcare. This paper aims to explore how existing frameworks conceptualize the influence of complexity in translating evidence into practice in healthcare. Data sources A database was interrogated using a search strategy to identify publications that present frameworks and models for implementation and improvement. Study selection Ten popular implementation and improvement frameworks were purposively selected. Data extraction Comparative analysis was conducted using an analytical framework derived from SHIFT-Evidence, a framework that conceptualizes complexity in implementation and improvement initiatives. Results Collectively the frameworks accounted for key concepts of translating evidence in complex systems: understanding the uniqueness of each setting; the interdependency of practices/processes and the need to respond to unpredictable events and emergent learning. The analysis highlighted heterogeneity of the frameworks in their focus on different aspects of complexity. Differences include the extent to which problems and solutions are investigated or assumed; whether endpoints are defined as the uptake of interventions or achievement of goals; and emphasis placed on fixed-term interventions versus continual improvement. None of the individual frameworks reviewed incorporated all the implications of complexity, as described by SHIFT-Evidence. Conclusion This research identifies the differences in how implementation and improvement frameworks consider complexity, suggesting that SHIFT-Evidence offers a more comprehensive overview compared with the other frameworks. The similarity of concepts across the frameworks suggests growing consensus in the literature, with SHIFT-Evidence providing a conceptual bridge between the implementation and improvement fields.