Collective scientific experience is a tool (…)

To prepare for the upcoming BorderLine meeting “Where Did the Experts Go?” And to start thinking about this subject from scientific experience, the Agricultural Science Expedition requires an analysis of many personalities. After François Saint-Pierre, it was the turn of Laurent Fleury, President of the Inserm Group Experience Pole, to take part in the exercise. Started in 1995, even before health expertise agencies were created, Inserm’s collective expert evaluations have earned their nobility letters over the years, combining researcher freedom with interdisciplinary analyses. feedback to understand both the strengths and limitations of this exercise, but also to question the difficulties of implementing such an approach in times of crisis.

Mission Agrobiosciences-INRAE. What, in your opinion, is the main quality of Inserm’s collective expertise?
Laurent Fleury: Let’s remember first of all that collective experience began in 1995, at a time when health expertise agencies did not yet exist. For Philippe Lazare, the general manager of Inserm who championed its creation, the researcher’s tasks are twofold: to discover and give advice or use his own words: “In the long term, allow for understanding; in the short term, help act on what we already know.”

In my view, one of the main qualities of collective experience is its multidisciplinary nature. Every day world news reminds us how important data from medical, clinical or basic research can be. The fact remains that these, by their number and variety, are difficult to access and use directly. This is particularly true in public health where the topics are complex, with factors that interact subtly, and issues that cut across with a strong social dimension. Its synthesis therefore requires a long critical analysis, which is exactly what Inserm’s collective expert reviews do.

Thus, for each of them, we ask the best specialists in the subject to critically analyze the available scientific literature, each of their own. For this purpose, we have a very efficient documentation service, with two seasoned notaries who make themselves available to researchers.

These multiple disciplines give expertise a great deal of power

These multiple disciplines give expertise a great deal of power. Of course, getting researchers from different disciplines to work together has never been easy, but that’s what makes the approach so interesting. In addition, in some cases this leads to new research questions or connections between people from different disciplines who find material to work together.

However, it has two drawbacks. First, this exercise requires time: it takes at least two years to implement a group experience. Second, it makes for a publication that certainly has the advantage of being accurate and detailed but remains difficult to read for the general public. Finally, I’d like to insist on the fact that I don’t like the term “expert”: it’s researchers who specialize in the respective fields we’re pursuing. We identify them primarily by their posts.

How specific are these experiences?

They are always about complex topics; Otherwise, we would not require collective expertise from Inserm. Moreover, and this is their second characteristic, we have great autonomy in conducting our investigations. Thus, the experienced shepherd intervenes at the beginning of the process, during the framing. Once this step is over, researchers are free to explore which items seem important to them, and what they particularly value.

Another peculiarity is that the researchers themselves write the document, which we then re-read, which does not give the same color as the document written by a third party, and is re-read and corrected by the researcher. We should also point out that we do not have a group leader, the experts in charge of the expertise are just facilitators who ensure and organize the smooth running of the expertise. Finally, Inserm can take hold of a subject that seems important to him. As far as I know, the institute has never confirmed this right. But its existence is a guarantee of autonomy, independence and freedom.

You distinguish between collective scientific experience and scientific experience in times of crisis, why?

Collective experience cannot be very useful in times of crisis: this methodology is more time-consuming than the response that a crisis requires. In addition, Inserm’s collective expert assessments are based on data published in scientific journals, which is not always possible, particularly on new or unknown questions.

However, some skills that are deployed during a collective experience can be mobilized in times of crisis, in particular the ability to identify people with expertise on a particular topic or even the ability to select relevant information from the scientific literature. In the time of the pandemic, our service has determined, every day, the available articles on Covid-19. Concretely, between April 1 and July 31, 2020, more than 29,000 articles were reviewed, including 14,500 selected and indexed according to eight topics of interest—epidemiology, psychology, immunology, and diagnosis, for example. Every Thursday, an abstract containing 35 to 40 relevant articles was sent to the ministerial unit responsible for crisis management. This information has been published via the Inserm website and sent directly to the scientists involved in the fight against Covid.

BorderLine meeting “Where Have the Experts Gone? Starts with the observation that scientific expertise is criticized and misused. Do you share it?”

I actually think there is a file A crisis of confidence From the public towards the scholars. This is fueled and exacerbated, in my view, by the social networks and media coverage of the “self-proclaimed experts” who come to give their personal opinions on the news channels continuously. Having said that, this phenomenon is not new, we have already been talking for about twenty years about the erosion of the credibility of academics or researchers. In this context, the media’s management of the Covid-19 crisis has not helped matters.

Our role is to provide expertise in a specific topic. Hence it is up to the politicians to make the decisions they deem necessary.

I have no real answer to give to this crisis of confidence. However, no matter which research institute conducts them – Inserm, IRD, INRAE ​​and more recently CNRS – collective expert assessments, because they are based on verified, validated and debated knowledge and not the opinion of an isolated person, are independent and reliable and have become tools of a healthy democracy. They are rarely questioned because it is difficult to come up and criticize a work based on a collective, interdisciplinary analysis. In addition, their presentation is transparent: all sources are clearly marked.

Our role is to provide expertise in a specific topic. Hence it is up to the politicians to make the decisions they deem necessary. Recently, expertise with pesticides has led public authorities to classify prostate cancer as an agricultural occupational disease associated with exposure to chlordecone. It’s a great success!

You are not in the same position as agencies like ANSES, which have the scientific expertise and are also responsible for marketing authorizations for phytosanitary products in particular.

It’s actually not the same role, the agencies have less freedom than us. Their experience is followed by a decision whether or not to allow the use of a substance. In addition, they do not have the same latitude that we do in selecting their scholarly article collection, which meet more stringent criteria.

In your opinion, are there blind spots in this topic of collective experience?

Instead of blind spots, I will mention two limitations or disadvantages. The first relates to the available publications. Our methodology is based on an analysis of the scientific literature. So it is necessary to have a certain seniority of subjects in order to get a sufficient number of search results. On the other hand, negative results are published less often, which may be a bias.

The second relates to the consideration of mundane or empirical knowledge, which remains difficult. Critical analysis of scientific literature is the work of researchers. This task cannot be entrusted to a person who is not familiar with the exercise. On the other hand, patients’ associations are dealt with, when possible, at the beginning of the process, when defining the specification of experience, to check, for example, that we have not omitted some points. When the process is completed, we present them with the conclusions of the experience. This approach works particularly well for associations that follow the same reasoning as us, in serving patients and in understanding pathology. This was the case for fibromyalgia, for example, where we were able to listen and discuss with different associations.

Interview with L.Gillot, Mission Agrobiosciences-INRAE, June 13, 2023.

[BorderLine] Where did the experts go?