Evidence-based policymaking in health systems is a myth
Is evidence-based policymaking "naïve and dystopian"? In this post, Devaki Nambiar (Research Scientist at the Public Health Foundation of India and a Member of the Social Science Approaches for Research and Engagement in Health Policy & Systems (SHAPES) and Translating Evidence into Action Thematic Working Groups of Health Systems Global) argues that for research knowledge to inform policy, it must be 'translated' by a range of other players who can make it 'culturally salient and institutionally viable'. This post originally appeared on the Health Systems Global website.
U.S. Army medical researchers take part in World Malaria Day 2010, Kisumu, Kenya April 25, 2010 (US Army Africa / Flickr CC BY 2.0)
Over the past several months, members of the Social Science Approaches for Research and Engagement in Health Policy & Systems (SHAPES) and Translating Evidence into Action Thematic Working Groups of Health Systems Global have scoured websites and interviewed a wide variety of practitioners to contribute to an inventory of knowledge translation initiatives linked to health systems.
There were many lessons to come from this exercise – which will be discussed at the upcoming Vancouver Symposium on Decoding Knowledge Translation: Initiatives, Institutions, and Perspectives on moving from knowledge to action and working across constituencies for policy/programme change.
Perhaps the most provocative of these was that evidence-based policymaking in health systems is a myth.
What is knowledge translation?
Obviously this was a revelation to us all. Isn’t evidence-based policymaking the central premise of knowledge translation? Not necessarily. Take the definition of knowledge translation arrived at during the 2013 Evidence to Policy Business Meeting of the Health Systems Research Symposium in Cape Town:
a situation where high quality, locally applicable synthesised research evidence, using the best available data informs, in the time available, action by elected officials, public servants, managers (including in NGOs), health workers, and patients/citizens (also donors, media and other intermediaries, including researchers)- systematically and transparently - in agenda setting, policy development, or implementation.
The verb in the definition is “informs.” This states the not insignificant distinction made in the health systems domain (as opposed to the biomedical/clinical domain) – that evidence is not the base of, but rather informs decision-making.
To many of the individuals I spoke to in building up this inventory – hailing from Uganda to Canada, Lebanon to Thailand, evidence-based policymaking is both naïve and dystopian. It is naïve in that it disregards the many other considerations that invariably shape decisions related to health systems, including the products, politics and passions by an array of players. These players are far reaching, from the citizenry to politicians, bureaucrats to researchers. It is also dystopian because the assumption is that this multiple stakeholdership, this pluralism is somehow less desirable than a situation that privileges evidence and the producers and purveyors of evidence, i.e. us!
The ‘architects and engineers’ of knowledge translation
Knowledge translation begins where research endeavours are often at their most vague. Research often determines the magnitude or mechanisms underlying health system challenges and problems. It sometimes compares intervention approaches to each other. But the question is most often, what ought to be done in this case? This is precisely where evidence can inform, but cannot serve as a base. With this in mind, knowledge translation is about other processes, institutions, disciplines even, converging in the service of better informed decision-making.
As was revealed to me interview after interview, this means that knowledge translation must involve a host of players – “architects and engineers” as one key informant put it, but also communication specialists, policy insiders, advocates and a range of issue experts whose engagement with each other has to be culturally salient and institutionally viable. Major knowledge translation efforts at various scales – global like the Evidence Informed Policy Network (EVIPNet), regional, like the Regional East African Community Health policy initiative (REACH), or national, like Thailand’s Health Intervention and Technology Assessment Programme (HITAP) – have fairly similar outputs and methods of engagement, but highly tailored, customised institutional structures taking this diversity into account.
The Knowledge 2 Policy Centre at the American University of Beirut, is a case in point. It is situated within an academic institution with a strong reputation and credibility. It has strong linkages to the McMaster Health Forum, also university affiliated, that has provided support and leadership for the EVIPNet consortium. Elsewhere, say Burkina Faso or Mongolia, similar initiatives have been based out of Ministries of Health. In Brazil, the Piripiri Evidence Centre was established by legal decree in 2010 at the subnational – municipal level. In yet other cases, such institutions operate autonomously – either with board membership of state Ministries of Health, as in the case of the European Observatory on Health Systems and Policies. Distinct and variegated as their institutional structures may be these institutions are very highly networked – with each other, but also with necessary stakeholders relevant to the scope of their work.
Impact: a slow trickle, not a revolution
The range of products for getting knowledge across to different audiences is as extensive as it is common across these institutions. These generally include policy or evidence briefs, summaries and notes, policy dialogues, outputs from rapid response services… and so on. However, when it comes to impact, “it’s a slow trickle, not a revolution.” It may not come as a surprise, but establishing networks, and remaining engaged with users of the research is what works.
What struck me, however, was that even this slow trickle takes place in a very specific context. It has to be set in an enabling environment that allows continuous advocacy. And the level of impact can vary or change over time. Nothing is set.
One informant described the four Ts essential to knowledge translation:
- Translation involving the use of language employed by policymakers (instead of research jargon)
- Tailoring – customising information based on specific questions policymakers may have or are focused on
- Trust – which knowledge translation practitioners have to painstakingly maintain – in the process trading the goal of evidence-based policymaking for evidence-informed policymaking
- Timeliness – understanding that policymakers are not always receptive and that the time has to be right for knowledge translation to hold meaning and advance an agenda.
Falling in love with an idea?
Across this broad range of topics on health systems, one interviewee argued that the process of knowledge translation is a constant negotiation, a dance whose moves keep changing. Where another interviewee explained: ’you need people to fall in love with the idea of evidence in policymaking.’
How do you make someone fall in love? Clearly – basing that process in evidence is inadequate!
We have concluded that there is both science and art, evidence and anecdote, strategy and serendipity involved in knowledge translation.
Provoked? Doubtful? Hopeful? Incensed?
Come tell us in Vancouver!
---
Devaki Nambiar would like to acknowledge the inputs of Preety Rajbangshi, Elaine Baruwa and Heather Cogswell to this blog post. This post originally appeared on the Health Systems Global website.