Defining and conceptualizing outcomes for de-implementation: key distinctions from implementation outcomes

Abstract

Abstract Background Increasingly, scholars argue that de-implementation is a distinct concept from implementation; factors contributing to stopping a current practice might be distinct from those that encourage adoption of a new one. One such distinction is related to de-implementation outcomes. We offer preliminary analysis and guidance on de-implementation outcomes, including how they may differ from or overlap with implementation outcomes, how they may be conceptualized and measured, and how they could be measured in different settings such as clinical care vs. community programs. Conceptualization of outcomes We conceptualize each of the outcomes from Proctor and colleagues’ taxonomy of implementation outcomes for de-implementation research. First, we suggest key considerations for researchers assessing de-implementation outcomes, such as considering how the cultural or historical significance to the practice may impact de-implementation success and, as others have stated, the importance of the patient in driving healthcare overuse. Second, we conceptualize de-implementation outcomes, paying attention to a number of factors such as the importance of measuring outcomes not only of the targeted practice but of the de-implementation process as well. Also, the degree to which a practice should be de-implemented must be distinguished, as well as if there are thresholds that certain outcomes must reach before action is taken. We include a number of examples across all outcomes, both from clinical and community settings, to demonstrate the importance of these considerations. We also discuss how the concepts of health disparities, cultural or community relevance, and altruism impact the assessment of de-implementation outcomes. Conclusion We conceptualized existing implementation outcomes within the context of de-implementation, noted where there are similarities and differences to implementation research, and recommended a clear distinction between the target for de-implementation and the strategies used to promote de-implementation. This critical analysis can serve as a building block for others working to understand de-implementation processes and de-implement practices in real-world settings.