Vous l’avez sans doute déjà repéré : sur la plateforme OpenEdition Books, une nouvelle interface vient d’être mise en ligne.
En cas d’anomalies au cours de votre navigation, vous pouvez nous les signaler par mail à l’adresse feedback[at]openedition[point]org.

Précédent Suivant

Chapter 6. PHIR, a process to support decisions and interventions

p. 121-147


Texte intégral

Social relevance at the heart of PHIR

1In the field of research in general, scientific relevance, along with methodological rigour, is a core evaluation criterion. It is a matter of showing that the research will improve knowledge on a particular topic. Thus, research teams need to demonstrate that they will not deploy their methods to describe a phenomenon that has been known about for decades (does smoking cause cancer?) or to test a hypothesis that is no longer original (do smoking cessation interventions mostly benefit the least poor?). PHIR being a form of research (see Chapter 1), it must necessarily satisfy the criteria of scientific relevance and methodological rigour (using appropriate and rigorous methods to answer questions). However, given its applied and contextual nature, it must also be ana-lysed in terms of social relevance. In other words, teams (see Chapter 5) that engage in a PHIR process also need to examine how their work serves the purpose of improving practices, actions, and policies, and, in some cases, even social change. While social relevance is at the heart of PHIR, we know from past reflections on the relationship between scholars and politics (Weber, 1963) that these interactions are neither obvious nor linear, let alone automatic (Parkhurst et al., 2018). A commonly espoused idea is that presenting scientific evidence will help decision-makers to act rationally (Ouimet & Bédard, 2015). Making decisions to change practices based on the results of one PHIR project is definitely not smooth sailing! Thus, the use of the research results does not come naturally. It may happen that political or strategic leaders take up PHIR results, but this is more the exception than the rule, as we saw during the Covid-19 pandemic.

2Teams embarking on PHIR must therefore become cognisant very early on of the importance of thinking about and planning activities that will be conducive a priori to the use of the results. As will be seen below, the state of knowledge on such activities is beginning to be widely available (Dagenais & Robert, 2012; Langer et al., 2016). Like intervention sustainability (Pluye et al., 2005), research results use must be carefully contemplated at the same time as the PHIR planning. The chances that the results will be both useful and used will be all the greater if this has been planned from the outset of the PHIR project and if specific and potentially supportive activities for their use have been budgeted and implemented. The final customary workshop for disseminating the results of a PHIR to decision-makers will have little influence on their decision-making if not accompanied by numerous knowledge transfer activities beforehand (Mc Sween Cadieux et al., 2017). One useful tool is the production of a knowledge transfer plan, at the time of PHIR planning, to be proposed upstream in the same way as an evaluation protocol or a scaling-up approach. Of course, over and above producing a knowledge transfer plan, forms of PHIR such as action research (Reason & Bradbury, 2001) or developmental evaluation (Patton, 2010) can also significantly increase the chances that results will be used, since this objective is intrinsic to their deployment.

3This knowledge transfer plan can be improved with the help of an expert in this field once the PHIR has been funded and launched. However, it can be developed during the drafting of the PHIR project with the collaboration of everyone involved, even if the researcher does not have all the skills to do so. In particular, it should specify:

  • what knowledge from the PHIR is to be shared: choose a priori the key messages and which specific dimensions of the study to emphasise;

  • the people and organisations to whom this knowledge would be addressed for decision-making or influence;

  • the knowledge transfer objectives, according the key actors;

  • who would organise the knowledge transfer process, and how;

  • when these activities should be implemented.

4Of course, it is never easy when starting a PHIR project or when just beginning to define its contours to know which knowledge will be most useful and for whom. Yet it is important to understand that producing a knowledge transfer plan is neither an end in itself nor an accountability document. It should be understood as a process to be developed progressively in response to PHIR implementation, needs, knowledge produced, windows of opportunity for putting forward results to inform decision-making, local and national contexts, etc. However, adopting a knowledge transfer posture from the outset and using a collective and participatory approach puts all stakeholders in the same boat when it comes to influencing decision-making with PHIR results. The knowledge transfer plan is not a panacea, but it is a tool that must be actively applied to fit the needs and be made useful. Finally, this knowledge transfer plan needs to be monitored and evaluated not only during the PHIR project, to adjust it based on the indicators to be followed, but also later on, to learn from our practices and improve afterwards. Reflexive practice, like social relevance, is a fundamental value of PHIR (Tremblay & Parent, 2014) and a core knowledge transfer competency, as we shall see below.

Processes that support findings use

5There are numerous and long-standing debates about the role of evaluation teams. From one extreme to the other, some (whether anthropologists or epidemiologists) believe these teams should limit their role to producing rigorous data on the intervention, whereas others argue that their involvement in supporting decision-making to improve actions is inherent to their profession. Michael Quinn Patton (2008) is famous, in particular, for having developed a school of practice: utilisation-focused evaluation (see Chapter 1). He recommends that the key question evaluation teams should ask sponsors and stakeholders is: What do you want to do with the results, and how will that help you? Use is one of the branches in the evaluation story tree (Lemire et al., 2020). For Patton, it is the conversation around the answers to these questions that will influence the evaluative choices and, by extension, how the PHIR project will be organised. While Patton does not mention the knowledge transfer plan, he proposes an evaluative process that is largely focused on it. Bamberger and Rugh (2012) later recommended basing these choices on four essential factors, all of which are constraints on the deliberative processes among PHIR stakeholders (see Chapter 5): time (of the research and intervention); data (availability, quality, and potential use, affecting the choice of approaches and methods, as presented in Chapters 3 and 4); budget (availability and size); and politics (issues related to decision and results use).

6This last element, therefore, repositions the issue of PHIR results use in the strategic and political arena, rather than confining it to a purely technical and methodological dimension, but still without pivoting towards the advocacy of activist associations. This does not mean, of course, that organisations with advocacy mandates cannot take up PHIR results or collaborate with PHIR teams to change practices and policies (Box 23). PHIR remains a form of research on public health interventions (see Chapter 1). As such, it clearly has practical and political implications – many public health journals now require authors to add a specific section on these implications – but the use of participatory action research from a social change perspective, such as for feminist or indigenous issues (Wehipeihana, 2019), is still (too) rare.

Box 23. An example of effective advocacy based on PHIR results

In Montreal (Canada), some associations have, for many years, exposed the injustice experienced by children of migrant parents without health insurance. The law in Quebec was such that these children, even when born in Quebec, were not entitled to receive care in public health facilities because their parents, often without administrative documents, were not insured under the public system. To report on these situations using a scientific approach, a research team worked with these associations to produce knowledge about the challenges and consequences related to healthcare access among migrants without health insurance (Belaid et al., 2020; Ridde et al., 2020). A partnership was organised, funding mobilised, and scientific and lay articles produced. The research team created knowledge-sharing videos on the situation, and a foundation put together a substantial advocacy file (https://tout-petits.org/publications/dossiers/acces-soins-de-sante-migrants) aimed at influencing legislators. The foundation mobilised all these resources and a very broad coalition of stakeholders to change the law. Finally, in 2021 the law was modified to give these children access to care; however, the law regarding pregnant women has not yet evolved, and the associations have mobilised again. There is still much knowledge transfer work to be done around access to healthcare for migrants in Canada (Merry & Pelaez, 2021).

7However, is it really accurate to say that these Quebec legislators used this scientific knowledge to make decisions? Might it not be more complex than that, as the Covid-19 pandemic has just reaffirmed? What uses and what knowledge are we talking about? What determines their choices or their ignorance (see Box 24)?

Box 24. The difficulty of determining what ‘knowledge’ means in public health

In managing the Covid-19 crisis, most countries convened one or more scientific committees to advise them on the measures to be adopted (Rajan et al., 2020). Almost everywhere in the world, these scientific committees or councils brought together infectious disease experts and clinicians, while failing to recognise that epidemic spread is above all a matter of behaviours and structures functioning as a system.
The measures thus implemented were aimed at one focal point (preventing hospital saturation) and on actions specific to these disciplinary sectors in response to the data used (i.e. if the virus is transmitted from one person to another through proximity and contact, then suppress contacts and proximity), with limitations on social interactions going as far as imposing lockdowns on entire populations and, to spur people to comply with these measures, anxiety-provoking communications and sometimes coercive measures.
These measures led to population exhaustion (pandemic fatigue), altered mental health, reduced access to non-Covid-19 care, and increased health inequalities. They also contributed to shifting a Covid-19 burden centred on seniors towards the younger population, on whom the measures had a major impact completely out of proportion to their vulnerability to Covid-19 (
Cambon et al., 2021). Confronted with this observation, the decision-makers defended their use of scientific data mobilised by their experts, without ever calling into questioning the ways in which they had chosen and used that data, nor how well the data represented the problem. Thus, in the first year, very little use was made of evidence about the measures’ acceptability over the long term, their negative impacts on overall health and on inequalities, the least detrimental communication channels, or the environmental risk reduction measures that could be applied (e.g. air purifiers and CO2 detectors); and in fact, this evidence was still hardly being mobilised in France in the second year.
This example shows how a mono-disciplinary approach can be harmful and can lead to detrimental health policy, especially when certain data (from this discipline) are chosen to be mobilised to the exclusion of others. Thus, using evidence is not a virtue in itself if the question is poorly asked, if the answers are not interdisciplinary, and if the modalities and processes used to respond to it are poorly defined. If those scientific committees had included other public health experts, such as specialists in prevention strategies, psychologists, sociologists, or stakeholders from the living environments, to consider less deleterious adjustments to the measures, the responses would undoubtedly have been different.

From results use to knowledge influence

8In the history of reflection on research use (Parkhurst et al., 2018), reference is often made to the work of Weiss (1979), who proposed three forms of use, which have since been widely disseminated. Thus, she suggested that research results use can be:

  • instrumental: to change practices, actions, policies;

  • conceptual: to strengthen the understanding of a phenomenon;

  • symbolic: to legitimise a decision, sometimes (often) already taken.

9This trilogy has given rise to many debates and refinements (Alkin & Taut, 2002), and the body of literature on the question of research use is growing (Cousins & Shulha, 2006). Some authors have since proposed not limiting the question of use to research results (findings use) but expanding it to include research processes (process use); they argue that PHIR stakeholders can also use research results while implementing an intervention (Cousins, 2008). This verges on Patton’s developmental evaluation approach (Patton, 2010), in which the evaluation team is not an outsider, but rather a stakeholder in the intervention. The team tries to ensure that the intervention is guided by the evaluation results, and that these results are shared as they emerge, without waiting for the long process of producing analyses and reports. This is, in a sense, what is referred to by the catch phrase implementation research, but over the life of the intervention and not only when it is being formulated, as proposed by the proponents of this concept (Peters et al., 2013).

10Thus, in a famous article, Karen Kirkhart (2000) proposed to advance the reflection on this question of research use by instead using the term influence, in the sense of the capacity of PHIR results to produce effects on decision-making through means that are indirect and thus, subtler than the sometimes instrumental and direct vision of the term use. By using the term influence, she proposed to widen the focus of research effects in a multidirectional (multiple sources), temporal (over the life of the intervention), or intentional (or not) perspective. The issue of non-use or even misuse has also been addressed in the literature on this topic (Cousins, 2004).

11Given that PHIR also aims to make knowledge useful for decision-making, this latter concept must also be examined. What knowledge are we talking about in the context of PHIR? Obviously, in the health sector, the reflex is to think of evidence, proof, and research results. This question of evidence is not new to PHIR, as it has been the subject of much debate, particularly among actors in health promotion (IUHPE, 2004; O’Neill, 2003), who have had – and still have – to fight to show that knowledge gained from their approaches, different from those of health sciences, evidence-based medicine (Sackett et al., 1996) (see Chapter 1), and other experimental methods, is equally valid to support and influence decision-making. This evidence can also be referred to as tacit or explicit knowledge, obtained not only from research, but also from evaluation or public health databases. For example, routine data collected daily in health facilities can be powerful information for evaluating natural experiments (Petticrew et al., 2005; Shadish et al., 2022) in the PHIR context using quasi-experimental designs (see Chapter 1) and time series. The effectiveness evaluation of a maternal health intervention funded by the Agence Française de Développement (AFD) in Chad recently showed the relevance of these routine data (Manoufi et al., 2021), even in a difficult context such as this country, which is regularly shaken by political turmoil. However, these data must also be accessible at the right time for PHIR, not to mention the challenges inherent in the ethics (and associated bureaucracy) of their use through requests for authorisation submitted to the administration and ethics committees.

12However, there is “no knowledge that is explicit in itself” (Catinaud, 2015). It is always contextual and influenced by social issues, as has been clearly demonstrated for health policies in Africa, for example (Fillol et al., 2020). Research results use and knowledge transfer are necessarily part of this contextualisation of evidence. Thus, beyond the knowledge derived from research, it is also essential to understand and consider the experiential knowledge and know-how of public health workers, which is not always systematically uncovered, even though it should be equally taken into account.

13Finally, the political analysis of knowledge production and of the epistemic and cognitive injustices it can generate should not be neglected (Piron et al., 2016). PHIR processes, in which power issues can be important, must necessarily take this question into account, especially (but not only) when working with populations living in difficult situations, such as migrants without administrative status, racialised people1, or indigenous populations, to take only these examples. PHIR must also become more inclusive. “Nothing should be done for them without them”, as Émilie Robert and her colleagues have proposed (Robert et al., 2018).

14There are, therefore, many kinds of PHIR results, and their use in interventions or at the end of analyses to influence decisions and practices is neither simple nor obvious. Training for stakeholders in knowledge transfer is as urgent to develop as it is essential to institutionalise (Box 25). Indeed, research and experience have shown that results use can be influenced by multiple factors (Dagenais et al., 2013), as summarised in the next section.

Box 25. Free online knowledge transfer training courses (in French and English)

The Renard team in Quebec (https://www.equiperenard.org), a research partnership team on knowledge transfer, in collaboration with the Institut de recherche pour le développement (IRD) in France have developed three free online courses in French. In response to the growing demand for knowledge transfer training for PHIR stakeholders (see Chapter 5), they decided to roll out these courses on a large scale to make them accessible to as many people as possible in the French-speaking world. The content of these online courses has largely inspired this chapter. Three courses are now available: introduction to knowledge transfer (20 hours); producing policy briefs (15 hours); and knowledge brokering (15 hours).
1. See:
https://catalogue.edulib.org/en/cours/renard101-en/ for the english courses.

Factors that influence findings use

15The strategies implemented to promote the use of PHIR results are one of the key factors influencing results use and will be discussed in detail in the next section. Here, however, we can stress the importance of choosing the most relevant strategy in line with needs and windows of opportunity; a morning coffee with a minister is sometimes more useful than a three-day workshop with 300 people…but not always. Also, regular, sustained, and direct contacts need to be established with potential users of PHIR results over the long term. This requires tenacity and resilience. It is also important to adapt the knowledge produced, not only to promote its uptake, but also to be able – which is never easy – to make it operational so that it is potentially usable. Finally (and there are numerous debates on this subject) some suggest that research teams should not confine their role to knowledge production, including in a co-construction process (Dupin et al., 2015), but continue to support decision-makers and stakeholders by accompanying them in the practice changes or policies they wish to implement. Thus, from this perspective, we can see how the current and traditional organisation of the research world (e.g. its methods of recruitment, evaluation, or funding) makes these processes difficult to implement (Ridde, 2009).

16Not all actors involved in PHIR (see Chapter 5) necessarily have the desire or the competencies to engage in knowledge transfer processes. In fact, there are many competencies, and a community of practice has proposed a list of essential ones (https://www.inspq.qc.ca/nos-productions/videos/referentiel-en-transfert-connaissances). This document sets out the knowledge required and the resources needed to conduct knowledge transfer and the behaviours to be adopted. These competencies are organised around five dimensions of a knowledge transfer process, akin to the processes deemed a priori effective according to a synthesis review carried out a few years ago on the science of using science (Langer et al., 2016):

  • adaptation of knowledge transfer knowledge and production of knowledge transfer tools;

  • knowledge dissemination and sharing;

  • support for the organisation of practices and knowledge appropriation;

  • support for the evaluation of knowledge transfer products, activities, and strategies;

  • support for the development of organisational capacity for knowledge transfer.

17The tool then proposes six essential competencies that are further subdivided into tasks within the five dimensions:

  • establishing and maintaining links and collaborations among stakeholders;

  • identifying knowledge transfer needs;

  • planning and managing knowledge transfer projects;

  • developing specific and relevant tools;

  • collecting, analysing, and synthesising information and evidence;

  • writing and presenting knowledge in an accessible way;

  • adopting a reflexive approach.

18Continuing with these factors, it is clear that research teams, through their own characteristics as well as their attitudes and competencies, influence results use. There are countless researchers who do not wish to get involved beyond the production of scientific articles, and arrogant professors who are unable to speak and communicate without jargon (from the use of the concept of epistemology to that of the odds ratio) and are complacent towards workers in the field, who are often eager to contribute to the research (Dagenais & Ridde, 2020). Humility is a rare competency in research (The Lancet Global Health, 2021), but it is fundamental to navigating and constructing with PHIR actors (see Chapter 5). Because PHIR is interdisciplinary in nature, the spirit of openness to other scientific disciplines and methods is paramount. In fact, in order to obtain data to share in this field, one must be able to study interventions in their totality. For example, there are sometimes endless debates between statisticians and sociologists about the nature of causality or evidence in PHIR. Without pragmatic compromises and decisions, it becomes difficult to propose relevant specifications (see Chapter 3) for evaluating the effectiveness of interventions. In one region of France, researchers wishing to test the heuristic value of the concept of mechanism (Lacouture et al., 2015) in realistic evaluation (see Chapter 3) around a PHIR project never managed to agree, as the sociologists and epidemiologists had a different view of the concept than the public health and evaluation experts (Breton et al., 2017). Researchers also need to be supportive of PHIR, so they should be interested in the use of rigorous methods, but not exclusively. They should be willing to venture into research to answer relatively operational questions. While theoretical, conceptual, or methodological analyses are essential to science, they are generally less of a priority in PHIR. As a result, research teams must be willing and able to maintain regular contact with implementers and decision-makers and ready to go outside of both their scientific comfort zone and their ivory tower (even if this image remains a myth). Human skills are just as important as technical skills.

19With regard to organisations, let us first consider the research centres or universities where the above-mentioned teams work. Indeed, these teams need to feel, and actually be, supported in these PHIR practices and be recognised in terms of career advancement. This type of research is often difficult to fund, and the long time required to build partnerships (see Chapter 5) does not allow them to publish as much as they might if using available databases, nor in journals with the greatest impact factors. Without abandoning the rigour inherent in any scientific approach, the traditional criteria for evaluating research teams should take this into account, for example, by giving more consideration to the building of long-term partnerships, rather than to the quantity of publications; the writing of articles in collaboration with field actors; and the production of documents and tools other than articles in peer-reviewed journals, etc. Similarly, organisations responsible for implementing interventions must be open to research and give their collaborators time to work with the research teams. The research community’s involvement in public health intervention organisations sometimes entails different time frames, greater rigour, as well as immersion in activities that provide or collect data for PHIR. Consider, for instance, the availability required from stakeholders when research teams need data to assess an intervention’s implementation fidelity (Pérez et al., 2019), or from focus groups that researchers conduct with teams to understand the intervention theory or the implementation challenges. This time dedicated to the research could be perceived as lost to the intervention but as gained for the PHIR project, and thus ultimately for the intervention. These issues also apply to organisations that make decisions based on PHIR results. They also need to have processes and policies in place to make the most of the products of research and promote evidence-based decision-making, even if no one has any illusions about the real influence of science. However, science is sometimes useful, and we should not despair; rather, everything must be done to sustain it. Resilience is also a quality demanded of both researchers – the example of those working on social health inequalities is eloquent (Ridde, 2019) – and decision-makers. As the Covid-19 pandemic crisis has clearly shown, scientific culture is still not very deeply ingrained in the thought patterns of those at the heart of policy decisions in many countries.

20On the stakeholder side, the front-line actors – the real policy makers (Lipsky 2010) – are in direct contact with the people for whom the actions are intended, and their attitude towards the world of research and their willingness to break with routines are important factors in the use of evidence. Like decision-makers, they may be more or less supportive of research and have basic knowledge of how evidence is produced in the context of an intervention evaluation. If they see the research as an activity to monitor their actions, this could discourage their involvement in PHIR, or even any reflection on the value of science upstream, when formulating their interventions. Obviously, it is never easy for anyone to examine themselves or to be willing to innovate by changing their usual practices. Moreover, many people involved in interventions have had negative experiences with research teams that, for example, patronised them or never returned to share their analyses of the data, which had in fact been collected collaboratively.

21The organisations that fund the research or intervention also play an important role in the use of results. These funders are still too often in separate camps, with those financially supporting the interventions being reluctant to see too much budget allocated to research (sometimes even refusing it) and those focused on research not really understanding why the researchers are so close to the actions. These organisations need to better understand the role of PHIR; those focused on the intervention need to understand that science (not consultation) is useful in developing and evaluating interventions, and those focused on research, that PHIR is just as useful and rigorous as biomedical or clinical research (see Chapter 3 on the scientificness of research approaches). Until more PHIR-specific funding organisations are available, both types of funders must accept that researchers will request budget for knowledge transfer activities and that interventions will plan on expenditures to conduct research with their academic collaborators. Some research funding agencies are just beginning to accept these expenses and to request knowledge transfer plans in protocols submitted in response to their calls for proposals. However, they should understand that innovation is needed and that strategies cannot be limited to a dozen lines at the end of the protocol where research teams say they are planning a final workshop to disseminate results and publish scientific papers. Thus, a greater sensitivity to, or even requirement for, knowledge transfer on the part of funders could influence practices and the use of science in decision-making. Unfortunately, money is often, but not exclusively, an effective incentive.

22The Covid-19 pandemic has reminded us of the extent to which the media play a role in how research data are taken into account when making individual decisions (e.g. vaccination) or collective decisions (e.g. the treatment of government measures) (Cambon et al., 2021). The situation in this regard is very mixed globally, as some countries have an independent press with a few scientific journalists, while others have none of this, with a press that is sometimes under government orders (Box 19), or mainly driven by the impetus to produce announcements. In Senegal, the controlled trial of the meningitis vaccine in 2007 gave rise to numerous controversies among the public and researchers, revealing journalists’ lack of scientific literacy and an attempt at political instrumentalisation (Ouvrier, 2015). Elsewhere, the Covid-19 pandemic period saw a proliferation of epidemiology training courses to equip journalists to process information, because their role was so crucial. In France, physicians with no real competency in epidemiology made boastful predictions in the media to journalists, who did not have the competency required to discuss or question the validity of their statements, and some journalists even made their own predictive calculations2. Better research use thus requires, among other things, increased knowledge on the part of media actors and channels to better disseminate PHIR results. This does not mean resorting to commercial appeals, but rather, sharing knowledge with teams of journalists who need to be able to inform society fairly and rigorously. For example, in recent years, in public policy formulation, the emphasis has been on taking into account the weight of influence of commercial industries, to the detriment of population health (de Lacy-Vawdon & Livingstone, 2020). Since the analysis of tobacco companies’ strategies (Breton et al., 2008), in particular, we know how much these industries can manipulate or distort research results through certain media and journalists. We also know that social media can be powerful vectors for, or against, the dissemination of evidence and debates on “fake news” (in French, infox) (Box 26). For example, in France, during the first part of the Covid-19 crisis, it was social media that publicised the value of wearing masks, even though the official position was the opposite, probably to conceal information on the absence of a national stockpile. Many PHIR projects are now trying to work with social media influencers to reach certain target audiences – especially, but not only, the younger ones.

Box 26. Science as seen by the press in Mali

In Mali, a 2017 analysis of about 100 newspapers and 14 interviews with their journalists provided valuable insight into how science is treated by the press in this West African country. For example, in the 2,500 printed pages of 242 editions found on newsstands over a two-week period, only 101 articles used scientific data. Health and the economy were the two sectors where they were most used; in the health area, medical issues were the primary focus. On the other hand, in the political area, which was very prominent in the press galleries, science was almost absent (only 15 articles). The study showed that “half of the articles use the scientific material ‘badly’: data not presented or inaccurate, low level of analysis, no citation of sources”. It also showed that the journalists who wrote articles in which science was present and well understood were those with stronger initial and continuing education, regardless of the diversity of their backgrounds. Others often settled for official documents or those handed out by people associated with the events or organisations they had to cover, without bothering to check their accuracy or to seek scientific data elsewhere. Their scientific literacy was generally very limited, and they often confused the notion of science with mathematics, numbers, and other percentages. This is reminiscent of Canadian biomedical researchers’ inadequate knowledge about qualitative research, showing the universal nature of training needs! (Albert et al., 2008).
Source:
Escot (2019).

23Faced with these multiple factors of influence, we must not give up and do nothing. The heart of PHIR is its social relevance, as we have said. It is therefore essential to engage in processes and activities aimed at increasing the likelihood that PHIR results will be used.

Processes and tools to strengthen findings use

24Rather than describing tools or activities, it may be more advisable to discuss knowledge transfer processes, since these should be multidirectional, interactive, and nearly continuous. As emphasised in the definition at the beginning of this chapter, the word “transfer” does not assume that some people know things and others do not, or that some people should transfer their knowledge to those without that knowledge. Rather, the concept of transfer evokes the notion of discussion and the need for multiple interactions. In fact, research has shown that unidirectional processes are not the most effective, regardless of whether these involve moving from science to practice (science – push) or from practice to science (demand – pull). Although the state of knowledge still warrants much development on this topic, the interactive approach appears to be a promising compromise between these two long-standing practices (Landry et al., 2006) to highlight the need for regular and sustained collaboration between research teams, decision-makers, and implementers (and other actors) for better use of PHIR results. The analysis of a knowledge transfer process aimed at using research results to reduce the consequences of road accidents in Burkina Faso resulted in five recommendations (Dagenais et al., 2021):

  • produce research data useful to stakeholders in the field;

  • ensure the acceptability of the technologies used for data collection;

  • use collaborative approaches for research and knowledge translation;

  • give visibility to actors in the field to provide them with more effective mechanisms for action;

  • involve high-level decision-makers more closely in the process to maximise the impact of the research.

25However, it is not easy to navigate these interactive approaches, as there are a plethora of conceptual models and frameworks attempting to explain how to support the transition from research to action (Graham et al., 2006). For example, the Institut national de la santé publique du Québec (INSPQ) proposes marking out the knowledge transfer process in seven steps (or sub-processes, for those averse to linearity): knowledge (co)production, adaptation, dissemination, reception, adoption by users, appropriation, and ultimately, use. The INSPQ makes a clear distinction between knowledge dissemination strategies and knowledge appropriation strategies and notes that the effectiveness of the approach depends essentially on the interaction, throughout the process, between research producers and users (Lemire et al., 2009). It is here that the question of the people or organisations at the interface of these two worlds must be raised, because very often these are two peoples inhabiting different planets. As such, it may seem difficult to organise any interaction between these two environments, and it may be more effective to use intermediaries.

26However, these intermediaries can be multidimensional, and navigating them is never easy. Thus, some authors recommend organising these functions along a continuum ranging from an informational role to a more systemic function within an organisation, to collaborate and build stakeholders’ capacities through a functional relationship in order to connect the PHIR actors and strengthen co-production (Shaxson et al., 2012). In this model, four types of intermediaries (Neal et al., 2021), not mutually exclusive, could be involved in these processes depending on the proposed approaches:

  • information disseminator: facilitates access to search results (informs, compiles data);

  • knowledge populariser: helps people understand and apply search results (shares, translates, communicates);

  • knowledge broker: improves the use of research results for decision-making; facilitates the co-production of knowledge (creates links, organises and supports meetings and interactions, networking);

  • innovation broker: influences the context more broadly to facilitate innovations (negotiates, develops, collaborates).

27The third type of intermediary, the knowledge broker, is increasingly discussed in the knowledge transfer literature (Munerol et al., 2013; Ridde et al., 2013) and is not new (Meyer, 2010). It has been the subject of experiments and evaluations to assess its effectiveness and explain its implementation challenges, in both North America and West Africa (Dobbins et al., 2009; Mc Sween-Cadieux et al., 2019). Indeed, knowledge brokering is certainly an intermediation profession of the future (Neal et al., 2021) in the world of PHIR, given that the two worlds, as mentioned above, are not always located in the same galaxy and that the people or organisations at the interface can undoubtedly contribute significantly. This person (or persons, within an organisation) can also engage in research if they feel they have the time and the skills for it. Some even feel they would have more legitimacy than non-researchers, which recalls the challenges related to humility in research and the ability to collaborate with a knowledge broker, thereby leaving research teams time to focus on the more scientific functions for which they have been better trained. This person will have to: assess the knowledge needs of potential users of PHIR results; identify and synthesise knowledge, and make it available to better develop intervention content and promote research use; organise regular meetings among PHIR stakeholders and create networks by playing, in particular, the role of liaison and exchange agent; strengthen capacities in knowledge transfer for stakeholders, but also more specifically, for example, in research methods for decision-makers and scientific communications for researchers; etc. These tasks will be multiple and varied, and will evolve according to needs and contexts. It is thus essential to find people for this role who have not only basic scientific knowledge, but also, if not more importantly, the social skills that are essential to the role of intermediary. Clearly, this person’s credibility with potential users is important, and it is never easy to find someone with all these qualities. These brokers obviously cannot claim to work alone, and they should be able to collaborate with researchers who are more expert in the subject and therefore more legitimately able to discuss it. However, we know that research teams struggle to find time for these interactions, as they are so taken up by their profession (research) and… administration.

28In one pilot project, the knowledge broker, who had been trained and mentored for two years by knowledge transfer experts, threw in the towel, largely because the Ministry of Health officials afforded him little time, space, and legitimacy. They considered that his master’s degree in sociology and his research training did not make him as credible, in conducting public health knowledge transfer, as a physician would have been, even one without a scientific doctorate. In fact, one minister of health understood the importance of these issues and created a knowledge mobilisation support unit. However, he was compelled by the regulations to recruit physicians, who were, in the end, neither trained nor interested in these knowledge transfer issues and who never got involved in setting up this unit. Thus, it continued to exist, but only on paper, and with the change of minister it was ultimately forgotten (Dagenais, 2021). Clearly, issues of power (symbolic or not) and context must be considered just as much as the budgets that should ideally be allocated to these intermediation functions. However, if the rare pearl cannot be found, it is conceivable to work as a team.

29Unfortunately, the broker described above did not have the support of the senior public health physician who was supposed to help him in his duties. The latter did not wish to get involved in this brokering with high authorities, no doubt so as not to compromise his lucrative consulting activities in an environment where criticism is a delicate matter (Olivier de Sardan, 2011). In this case, the knowledge broker was able to mobilise the researchers’ scientific credibility in a specific and ad hoc way when his analysis of the context and the timing indicated that it was relevant to do so. An example of brokering activities and the qualities required for this person is presented in Figure 12. A systematic review proposed a list of 10 potential knowledge brokering activities around three main functions: knowledge management, exchange and link management, and capacity building (Bornbaum et al., 2015).

Figure 12. Example of knowledge brokering activities and qualities required.

Image 10000000000004A700000347C18CD3BFDE304BF0.png

Source: Mc Sween-Cadieux (https://ideas4development.org/en/research-development-knowledge-brokering/).

30Given the many possible knowledge transfer activities, which can sometimes be specific to particular contexts (Siron et al., 2015), a research team conducted a mammoth review to understand the different levels of effectiveness (Langer et al., 2016). From that review, they proposed a list of six a priori effective mechanisms for knowledge transfer:

  • raising awareness and fostering positive attitudes towards the use of evidence;

  • building consensus on and adapting knowledge about PHIR issues relevant to those who will make decisions;

  • strengthening communication and access to evidence;

  • facilitating interactions among decision-makers and research teams;

  • developing skills for accessing and understanding evidence;

  • influencing decision-making structures and processes.

31In a recent review, another team identified no fewer than 38 strategies that would be effective for using research results (Zhao et al., 2020). Finally, as part of a PHIR project conducted in France, a taxonomy of knowledge transfer activities was created and assessed as relevant for deployment when implementing prevention policies in local settings. Thus, 18 standardised knowledge transfer activities (each with a specific objective and wording) were produced, grouped into 11 categories of activities (Affret et al., 2020).

32We will spare you these long lists, which are accessible in the two articles. Beyond these processes and activities, there are also specific and useful tools that cannot all be described in this book: infographics, cartoons (see Box 27), videos, websites, communities of practice, etc.

Box 27. Using cartoons to share search results

In Burkina Faso, a team of researchers, helped by a knowledge broker and members of an association that organised an intervention being analysed as part of a PHIR project on the elimination of user fees for children’s healthcare, partnered with a cartoonist to disseminate the results. The use of cartoons was one knowledge transfer activity among the many others organised by the team (workshops, policy briefs, website, press, etc.). However, the cartoons were a very innovative device, because the idea was to share serious and rigorous research results, but in an accessible and humorous way.
A great deal of work was done upstream to ensure that the researchers explained the research results so that the graphic artist could assimilate the content for cartooning. A researcher and a knowledge broker worked extensively with the cartoonist, particularly on several drafts, to ensure the scientific and contextual relevance of the content presented. The intervention stakeholders were also asked to verify the social acceptability of the drawings, as humour is obviously cultural. The collaboration with a cartoonist who is well known in Burkina Faso and West Africa greatly enhanced the relevance of the proposals and their dissemination. Humour made it possible, on many occasions (but not always), to spark debate on the PHIR results. For example, the idea that eliminating user fees would cost too much for the State to subsidise was illustrated by comparing this cost with the cost of three beers. These cartoons were used in an album that was both in print format and online
1, and was presented at press conferences, national workshops, and international conferences.
1. See:
https://www.acfas.ca/publications/magazine/2015/04/caricatures-partage-savoirs-contrer-oulipo-academique

Figure 13. Financing access to health services for children in West Africa costs less than $5 per child per year and requires the political will of decision-makers.

Image 10000000000004A50000034813497E653D98DB0E.png

© Damien Glez

Figure 14. Using caricatures as a tool for knowledge transfer.

Image 10000000000004A5000003487E3B39C30AFD8717.png

© Damien Glez

33However, while there are numerous knowledge transfer tools, policy briefs and deliberative workshops are innovative ways that definitely should be used more widely.

34Policy briefs are short texts written in plain language and presented in an attractive format (Dagenais & Ridde, 2018). A brief summarises the research results and makes operational recommendations to decision-makers and stakeholders, proposing solutions identified in the PHIR project to improve practices, interventions, or policies. It differs from research summaries and other research notes or research snapshots in that its purpose is not only to present PHIR results, but also, if not above all, to make practical recommendations to inform decision-making, for example, at a deliberative workshop (see below). The brief can be written in the format presented in Box 28. It should serve as a basis for dialogue with stakeholders interested in or affected by an issue. It is essential to take the time to prepare and write this brief well in advance and to test its content and form with a sample of target individuals to ensure that the recommendations are relevant and the language level is appropriate. Otherwise, there is a risk that on the day it is used – for example, during a deliberative workshop – a senior official who was not involved upstream will dwell on details of form (the title is too punchy, it calls into question my agency, etc.) only to denigrate the substance (which they will not have read) because the PHIR results will not correspond to what they wanted to hear or do.

Box 28. Policy brief design

Size: Maximum 4 pages printed on both sides; 1,000 to 1,500 words depending on the use of images.
Title: Short, punchy, informative.
Summary: Persuade the reader to keep reading.
Highlights: 3 or 4 messages in a box.
Introduction: Explain why this topic is important, why the reader should care, what the objectives of the PHIR were.
Approaches and results: Summarise facts, background, and available data; reduce details to what the reader needs to know about the PHIR project; provide concrete examples to support your claims and operational recommendations.
Conclusion: Base it on the results presented; provide concrete conclusions and supported claims.
Sources consulted or suggested: Research report or articles on which the brief is based (web link).
Recommendations for action: The key measures or actions you suggest (to be done by whom, when, where?) that are realistic and feasible.
Source:
Dagenais & Ridde (2018).

35This dialogue can be held particularly during deliberative workshops; it is recommended that policy briefs be distributed beforehand and analysed during the workshop to generate dialogue among PHIR stakeholders (Boyko et al., 2012; Mc Sween-Cadieux et al., 2018). These workshops are defined as: “an evidence-informed, deliberative dialogue process among multiple stakeholders for vigorous and comprehensive policy and practice decision-making” (Nabyonga-Orem et al., 2016). At its heart, this is an interactive approach that brings together all those affected by the problem being addressed by the PHIR project and considers that collective intelligence will be more effective in finding solutions than the intelligence of a few people. As we attempted in Niger (Hamani Souley et al., 2017), for example, the aim is to establish dialogue and deliberation, not debate and confrontation, as is all too often the tendency in the French-speaking world. An environment is needed that will allow stakeholders to collaborate and to discuss the contents of the research results in light of their own empirical knowledge, so they can understand and learn together to make informed choices using PHIR. For this, it is sometimes necessary to call on people with real group management and professional facilitation skills to handle the power issues inherent in these modes of group consensus-building. Based on our collective experience in organising these deliberative processes, we propose in Box 29 some lessons learned for those who would like to engage in these processes.

Box 29. Lessons learned from conducting deliberative workshops

- Vary the types of data presented from practical experiments and both qualitative and quantitative research.
- Ensure the presence of a diverse group of stakeholders concerned by the PHIR project.
- Provide advance notice and prepare participants for what is expected of them during the workshop.
- Make data and presentations viewable and accessible (form and background).
- Allow time for deliberation, debate, and discussion.
- Create small working groups to explore certain topics in greater depth.
- Produce operational recommendations that are solidly grounded in research and experiential data.
- Take into account power issues in the preparation, organisation, and follow-up.
- Do not turn the workshop into an advocacy exercise for a particular project or solution.
- Prepare the workshop in advance: prior contacts with decision-makers, presenters, policy briefs, etc.
- Prepare data summaries in the form of briefs, in appropriate language, and distributed (and if possible tested) in advance.
- Allow time at the end of the workshop to clarify the content of the recommendations and present them to everyone before the end.
- Set up a committee to follow up on recommendations and give it the means to function.
- Assess the processes and outcomes of the workshops in order to improve.
Source:
Ridde & Dagenais (2017).

Links and conflicts of interest during PHIR

36To conclude this chapter, we need to address a sensitive issue rarely discussed in public: research independence and links/conflicts of interest in the context of PHIR.

37As we have seen in the previous chapters, PHIR is a process that involves multiple actors from different backgrounds and cultures, but which is based on the principle of associating the practice of science with that of action. There are, therefore, significant risks that stakeholders will seek to focus the research on what works well (see Olivier de Sardan’s [1995] concept of “enclicage”), rather than on the challenges encountered during the intervention, or that the funders will push research teams to show effectiveness where there is none Gorman, 2018). These issues become crucial in settings that are resource-poor or dependent on international aid or national public funding (see Box 23). This is especially true when funders want to disseminate “ready-made” intervention models without evidence of their effectiveness or adaptability to local contexts (Gautier et al., 2020; Olivier de Sardan, 2021). Similarly, the populations and individuals benefiting from the intervention or its advantages do not always look favourably upon a research team that might call into question this presumed usefulness or effectiveness (see Chapter 2). The “success cartel” is on the prowl (Rajkotia, 2018). For example, some researchers have recently come forward in public health journals with calls for more vigilance around research independence (Storeng et al., 2019) in a context where public funding is increasingly competitive and drying up.

38In a PHIR project, there can, in fact, be multiple influences, and conflicts of interest cannot be reduced simply to the familiar ones from the pharmaceutical, tobacco, alcohol, or agri-food industries. What about links of interest when, for example, the director of an association that negotiates with and then receives significant funding from an international organisation to carry out a PHIR project becomes a member of that organisation’s management team a few months later? What about a group of global health researchers who say they want to independently influence their country’s policy, when they accept funding from that country’s development bank and hold their meetings in the offices of a research funding agency in that country? Researchers and evaluators in the field of intervention evaluation are obviously also affected by this (see Box 30).

Box 30. Two examples of attempts at influence by a PHIR funder

A development bank carried out an impact evaluation of an intervention for more than 2 million USD. It funded both the entire intervention and the entire evaluation. Its own contracted experts co-authored several scientific papers based on the evaluation. The inconclusive results had not been made public more than two years after the end of the intervention. Yet they had been known for a long time. No official final results workshop was held, and the funder never wanted to organise one. A workshop to share the preliminary results did take place, but no document was given to the participants and they never had access to the slides presented. This strategy should no doubt be attributed to the fact that this bank was negotiating with the government to convince it to continue this approach more than 18 months after the end of the previous project. The report was finally made available online more than two years after it was written.
A public health physician supported an intervention as a consultant for many, many years. One of his first evaluation reports on its effectiveness was quite positive, although the methodology used was not rigorous enough to support such a claim, which no one in the organisation questioned. Later, the funder commissioned researchers to carry out an independent evaluation. It appointed one of its employees, who was also trained in research but did not have an academic position, to follow up on the matter. As time went on, she interfered increasingly in the thinking and methods proposed by the researchers. She became more intrusive, correcting all the details, questioning all the solutions, and, above all, looking for statistical solutions so the evaluation results would be positive. However, nothing worked; the intervention was really not very effective despite the millions invested. A little later, new data, after multiple attempts, showed a positive but relatively weak result. A workshop to plan the next stage of the intervention was organised, to which the researchers were not invited. The consultant attended, and denigrated the results of the research on “his” intervention. Locally, no one dared to question the sustainability of the intervention, which continued, and the consultant was again called upon to support the process. This did not prevent the funder from later stating in a book (coordinated by its employee!) that the work of the research team had been useful and was taken into account to radically change the intervention.

39This type of conflict of interest was studied a long time ago by Scheirer (1978), who discussed cognitive mechanisms that could push evaluators to emphasise the positive effects of actions. She cited a review of the 1970s literature showing the challenges of internal evaluations (which have since been largely abandoned in favour of external evaluations) since “evaluators who were affiliated with the evaluated intervention were much more likely to report the success of the programme (58%) than non-affiliated researchers (14%)”. There are no magic recipes for dealing with these conflicts or links of interest, but it is essential to talk about them as a team, among PHIR stakeholders, and to make them the subject of calm and dispassionate discussion to find appropriate solutions. This is also recommended when deciding who signs (or not, and in what order) scientific articles resulting from PHIR projects. It is also important to be transparent and honest outside of these groups so that readers of articles that will be published as a result of the PHIR project, or who will listen to a conference presentation on these results, will understand the issues at stake and analyse the results in light of these links of interest.

Conclusion

40The purpose of Chapter 6 has been to make very clear the importance of knowledge transfer in the field of PHIR. The use of PHIR results is an important criterion of effectiveness. Besides deploying rigorous methods to strengthen the credibility of the results, it is essential to plan and organise specific activities that will increase the potential for their use.

41Of course, nothing is easy, and use is never systematic or straightforward. It is mainly a matter of trying to influence decision-making, as there are many constraints on knowledge transfer, whether organisational, institutional, personal, contextual, or cultural. With this in mind, we have provided an overview of knowledge on the effectiveness of certain interventions and tools that should inform your choices in order to strengthen the use of your research.

Notes de bas de page

1 Simona Tersigni explains that “a racialised person is subject to a process of categorisation and differentiation according to hereditary somato-psychological characteristics socially instituted as natural” (Crenn & Tersigni, 2012).

2 See: https://www.lemonde.fr/planete/article/2021/06/18/le-lourd-cout-humain-d-un-troisieme-confinement-tardif-en-france_6084619_3244.html

Précédent Suivant

Le texte seul est utilisable sous licence Creative Commons - Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International - CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.