Integrated Knowledge Translation for Implementation Science

A pressing concern for those concerned with implementation is how best to engage with stakeholders.

Anita Kothari highlights learnings, tensions and challenges from a Canadian approach designed to foster working with stakeholders to promote the implementation of research findings – a critical element of many implementation and improvement projects.

Integrated knowledge translation (IKT) is a model of collaborative research where researchers work with knowledge users who identify a problem and have the authority to embed the research recommendations. Perhaps slightly differently to research co-production, in IKT the emphasis is placed on shared decision-making during a research process. The focus is fixed on implementation of findings. Other characteristics of IKT include:

  • joint refinement of the research questions and design; questions reflect needs from the practice/policy setting but are also of interest to the researcher

  • knowledge user involvement in data collection, where possible, and joint data interpretation

  • sharing of expertise (rather than trying to turn knowledge users into researchers)

  • sharing of perspectives about the problem and the context, leading to a shared understanding of possible solutions

  • joint development of a knowledge dissemination and implementation plan

  • joint authorship of knowledge products resulting from the research

IKT is an ideal approach to implementation for several reasons. A better understanding of the context for implementation is surfaced during the collaborative process. In my experience, knowledge users are eager to discuss the program or policy legacy of similar initiatives, providing clues as to previous implementation barriers and facilitators. Knowledge users may be able to provide details about the current state of capacity to deliver new programs, or the ability to gain support for new policy direction. This is helpful in determining whether front line practitioners require additional skills development to deliver the new program effectively. Perhaps other resources, like IT support, are required to move the research findings forward. Finally, knowledge users can assist with aligning new programs or policies with current organizational priorities (or future ones). In short, an IKT approach leads the way for feasible implementation.

The first set of tensions is related to the way in which we think about research, knowledge and evidence:

Below I offer some observations to launch further discussions about the role of IKT and implementation science. These comments are based on my extensive experience conducting or participating in IKT studies, mostly in the area of public health and policy making. In addition, colleagues and I are conducting a realist review of IKT studies to determine what arrangements work, for whom and under what conditions. Using this foundation, I asked myself: to what extent is IKT a promising implementation approach? This reflection led to the identification of tensions or challenges that beg further discussion.

In my experience, there is an expectation that the work of collaboration will unquestionably end with study findings that warrant change (implementation). That is, research collaborators expect that the findings will be in a positive direction or large enough to require practice or policy change. There is no reason, however, to make this assumption, and we don’t talk about this very much. Both knowledge users and researchers need to be concerned about this bias before initiating a research partnership given the resources required to support a successful collaboration.

We also don’t question whether using an IKT approach gives us a better scientific process. In thinking about this further, it’s not even clear how we would measure the IKT process to determine if it is scientifically more robust than a traditional approach. Do we turn to indicators of validity and generalizability? Credibility? Social robustness? While both researchers and the scientific community ought to be interested in this question, funders who fund IKT research arrangements need to be examining this question more closely.

The second set of tensions is related to the role of research partners:

Finding yourself in a productive IKT collaboration is rewarding on many levels. One of the indicators of a successful partnership is what I call, ‘the asks’. Your knowledge user partner may call you to ask for a few articles about such-and-such topic, or ask whether you might do a presentation to a different department than their own. (The researcher ‘ask’ will likely be to partner on a future project). Both researchers and funders ought to be aware of these additional requests that occur outside of the research project. Funders might go so far as to put aside funds or ask research proposals to include funds to address the ‘asks’ – they represent real-time problems that are looking for research input.

Researchers and funders ought also to be concerned with the power dynamics in an IKT relationship. Unlike participatory action research, IKT approaches do not strive for empowerment. Nevertheless, issues like the co-opting of the research direction by one of the research partners, or determining how decisions will be made throughout the research process, require further exploration to move IKT processes to acceptable implementation. (We are currently writing a paper to tease out issues related to power and IKT).

The last set of challenges relates to how decisions are made:

It is often the case that my research collaborators want to share the on-going findings with their department. As a researcher, that makes me supremely uncomfortable. How far along in a, say, five-year study can we do this with confidence? How do we present the findings in a tentative yet acceptable way? Researchers and collaborating departments need to be aware of the limits of preliminary findings. More importantly, the scientific community hasn’t given enough guidance to researchers engaged in applied research about how to do this with confidence and robustness.

My last observation is about activating knowledge user networks to promote the uptake and implementation of research findings. IKT scholars have focused on knowledge user partners as the prime users of the co-developed research, but these knowledge users have far-reaching networks where the research findings might also be applicable. More energy is required from researchers, and perhaps knowledge users themselves, to determine how to activate these networks for implementation of findings with broad reach.

The focus in this blog has been on sharing power and decision-making as part of the knowledge translation and exchange process. This is a critical element of the wider process of implementation of change and improvement where learning needs to be shared across boundaries. I welcome further thoughts about these observations. How do your experiences with IKT and implementation align with these reflections?

Anita Kothari is Associate Professor at the School of Health Studies at the University of Western Ontario, Canada. Her research interests focus on knowledge translation and exchange, and public health. She gave a guest lecture open to UK-IS members at Kingston and St George’s University of London earlier this year.

Anita Kothari, Western University (

© UK Implementation Society, 2018

All views expressed are the author's own and not those of the UK Implementation Society.

Recent Posts
Search By Tags