Knowledge translation in health: how implementation science could contribute more

Wensing, M., Grol, R. (2019) Knowledge translation in health: how implementation science could contribute more. BMC Med 17, 88. https://doi.org/10.1186/s12916-019-1322-9

Abstract

Background: Despite increasing interest in research on how to translate knowledge into practice and improve healthcare, the accumulation of scientific knowledge in this field is slow. Few substantial new insights have become available in the last decade.

Main body: Various problems hinder development in this field. There is a frequent misfit between problems and approaches to implementation, resulting in the use of implementation strategies that do not match with the targeted problems. The proliferation of concepts, theories and frameworks for knowledge transfer – many of which are untested – has not advanced the field. Stakeholder involvement is regarded as crucial for successful knowledge implementation, but many approaches are poorly specified and unvalidated. Despite the apparent decreased appreciation of rigorous designs for effect evaluation, such as randomized trials, these should remain within the portfolio of implementation research. Outcome measures for knowledge implementation tend to be crude, but it is important to integrate patient preferences and the increased precision of knowledge.

Conclusions: We suggest that the research enterprise be redesigned in several ways to address these problems and enhance scientific progress in the interests of patients and populations. It is crucially important to establish substantial programmes of research on implementation and improvement in healthcare, and better recognize the societal and practical benefits of research.

This article calls for more and better research on knowledge translation and implementation science claiming “in recent years there has been little progress in our understanding of how healthcare practice can be improved.” The authors cite five main challenge areas faced by researchers studying knowledge translation and implementation science.

  1. Misfit between problems and approaches to implementation, resulting in the use of implementation strategies that do not match with the targeted problems.
  2. The proliferation of concepts, theories and frameworks for knowledge transfer – many of which are untested – has not advanced the field.
  3. Stakeholder involvement is regarded as crucial for successful knowledge implementation, but many approaches are poorly specified and unvalidated.
  4. Despite the apparent decreased appreciation of rigorous designs for effect evaluation, such as randomized trials, these should remain within the portfolio of implementation research.
  5. Outcome measures for knowledge implementation tend to be crude, but it is important to integrate patient preferences and the increased precision of knowledge.

See table 1 for analysis of the challenge areas and suggestions for improvement.

Two reflections on these five and the accompanying analysis

#2: Proliferation of frameworks: I think it was Karen Ritchie (NHS Quality Improvement Scotland) who said “Frameworks are like toothbrushes. Everyone has one and nobody wants to use anyone else’s”.  I agree with the authors’ assessment that we have too many frameworks that lack rigorous assessment and application in practice. Most are conceptual frameworks which is nice as a concept but how does that help in practice? The authors state that more research on fewer frameworks “would advance science far more than the continuous development of new theories within disconnected worlds. Rather than adding new frameworks, the focus should be on the testing, refinement and integration of theories.”

#3: Stakeholder engagement in research. This is heralded as the “holy grail” of health care improvement according to the authors but there is little research on how to do this effectively. They cite a potential challenge when input from stakeholders contradicts available knowledge. The authors call for more research on stakeholder engagement. “We suggest that methods for stakeholder involvement must be better specified and validated in empirical research.

There is a good discussion of the failure of health research funding to recognize some of the unique challenges in research on knowledge translation and implementation. Similarly there is a reflection that the academy doesn’t easily support these types of research which is inherently interdisciplinary and advancement happens over a long time. “This research is best supported in networks that bring together scientists with different backgrounds who can work on sequential projects over a longer period of time.” This makes sense since models that cover engagement, research, dissemination, adoption, use and evaluation – like the Knowledge To Action Cycle or the Co-Produced Pathway to Impact – require different expertise working in different settings over time.

Questions for brokers:

  1. The authors state that involving stakeholders is a holy grail. If so, then knowledge mobilization/implementation practitioners should be included in research on knowledge translation and implementation, but this doesn’t happen often. Why not?
  2. The comments about health research funding do not mention the issue of peer review of health research applications. How do we redesign the health research enterprise (as called for in this article) if those who have the power to redesign it are also the ones who need to change or be changed?
  3. What needs to change in the academy to recognize and reward networked, interdisciplinary health improvement focused research over time?

Research Impact Canada is producing this journal club series to make evidence on knowledge mobilization more accessible to knowledge brokers and to facilitate discussion about research on knowledge mobilization. It is designed for knowledge brokers and other parties interested in knowledge mobilization.