Mircea Sofonea: "Modeling is a form of quantitative data analysis like any other, not a divinatory practice".

Mircea Sofonea is an epidemiologist in the Infectious Diseases and Vectors: Ecology, Genetics, Evolution and Control laboratory (Mivegec[i]). For almost a year now, his discipline has been front and center, and the Montpellier-based researcher has been particularly busy in the media. He looks back on a decidedly special year.

How did you get into epidemiology?

After studying biology at the École normale supérieure I defended a thesis in evolutionary epidemiology at the University of Montpellier in 2017. It's a city with an exceptional dynamic in the life sciences: from population biology to health. And epidemiology, the study of health determinants over time and space, is at the crossroads of many disciplines. The epidemiology of infectious diseases in particular requires the involvement of clinicians, microbiologists and public health physicians, as well as ecologists, evolutionary biologists and applied mathematicians... At Montpellier, I found a highly favorable environment for developing interdisciplinary approaches.

Has the occurrence of the covid-19 pandemic changed your work?

On March 12, 2020, most of the research carried out by the Theoretical & Experimental Evolution team headed by Samuel Alizon at the Mivegec laboratory switched to covid-19. Our team was one of the first to produce work and software for France in a public way: on April 6, we put online a simulator enabling us to estimate how many people had been infected, in order to determine if and when herd immunity would be reached, and when the peak in hospital occupancy would occur. As a result, we were one of the few French teams to bring our expertise in epidemic modeling to the national level, and in late April we were contacted by the FrenchParliamentary Office for the Evaluation of Scientific and Technological Choices (OPECST, chaired by Cédric Villani). We contribute to the dissemination of science, but we do not prescribe public policy, even if many media outlets are urging us to do so.

You were very much in demand by the media around covid-19. Were you familiar with this relationship with the press?

At first, the media and journalists were not aware of epidemiology, and we were contacted by word of mouth for our educational notes on the subject. Then the need to understand and quantify the epidemic translated into daily requests, mainly from the media, but also from health professionals and even the private sector. We thought the exhibition would last a few weeks, but almost a year on and over 300 interviews later, the requests are still coming in, and I'm having to turn them down due to lack of time to maintain my research activity in an unprecedented context of scientific production, as well as a full teaching load. I have also been able to observe the effect of a pause in the response to media requests: the observation shared by other colleagues is that scientific analysis quickly gives way to misleading discourse, which not only undermines public confidence in the scientific community, but also weakens the public's understanding of and support for health measures.

What do you think of the various, and not always concordant, comments we've heard?

Don't use urgency and media hype

to fill the void, and some scientists took advantage of this effervescence to assert unestablished truths outside their field of expertise. This kind of rhetoric ended up losing public opinion. Containment, for example, is a decision that cannot be taken on the spur of the moment, but must be backed up by quantitative arguments and appropriate scientific data and analysis.

We've read a lot of articles in the media in which you've been interviewed, headlined " We must confine ". Do you find your words echoed there?

This is one of the difficulties encountered when dealing with the media: we often find that the reproduction of what we say is not entirely faithful, and in the published article we find neither the conditional used nor the oratorical precautions, even though we are trying to convey a nuanced and balanced view of research, which is by definition exploratory. It's vital to distinguish science from research. The former represents a body of knowledge established and accepted as such by a specialist community, whereas research is a human activity like any other, with a way of producing it that is specific to each individual, from intuition to preliminary results, but not yet at the level of consensual fact. The media all too often mix the two.

How do you see your work evolving after covid?

I hope that we will be able to return to long-term research and contribute to a transdisciplinary scientific understanding of infectious diseases, as well as passing on this knowledge, in particular by promoting better health education for future generations. More generally, researchers into the epidemiology and evolution of infectious diseases are currently seen as the bearers of bad news, whereas many of these diseases are preventable, and investment in prevention upstream of the emergence and spread of an epidemic is crucial. There's a form of ingratitude in public health: if you foresee and anticipate an event, everything is done to prevent it from happening, and you're seen as a catastrophist. If you don't foresee it and it happens, you've done your job badly. It's important to remember that epidemiology answers 3 questions: understanding the past, describing the present and shedding light on the future, but not predicting it. Modelling is basically a form of quantitative data analysis like any other, and not a divinatory practice, as some observers like to call it, forgetting that behind every interpretation of the figures lies a model, however simplistic and implicit.

Had you experienced any other epidemics before covid-19?

With Samuel Alizon and two interns from the MEME master's program, we worked on the Ebola epidemic that occurred in West Africa from 2013 to 2016. This epidemic was unprecedented in its scope and duration, and it was important to study its long-term consequences. This is a good example of the interaction between public health and viral evolution. In developed countries, it was believed that antibiotics and vaccines would quickly solve the problem of infectious diseases, and that the focus should now be on chronic illnesses. The covid-19 epidemic proved us wrong.

[i] Mivegec : UM, CNRS, IRD