What do we know about the role of schools in the Covid-19 epidemic? Five experts answer

The question of school closures has been the subject of recurrent debate since the start of the Covid-19 epidemic. It is now clear that severe forms of the disease affect children only very exceptionally. But what about the capacity of the youngest children to transmit the virus? Is keeping schools open a problem? Can closing them be effective? Why is it so difficult to reach a consensus on the role of schools in the dynamics of the Covid-19 epidemic? To find out, we interviewed five experts.

Lionel Cavicchioli, The Conversation

"In our models, we do not consider that children have a different profile from adults.

Mircea Sofonea, lecturer in epidemiology and evolution of infectious diseases at Montpellier University

Assessing children's contribution to the dynamics of the epidemic has been a crucial issue in our understanding of the epidemic from the outset, but it is fraught with methodological difficulties and legitimately raises the question of the cost-benefit balance of measures targeting the school environment. On this very subject, not all studies have come to the same conclusion. Some French references initially downplayed this impact, before the results of theInstitut Pasteur's ComCor study, while German and British references were more pessimistic. According to the Germans, viral loads are as high in children as in the rest of the population, a fact confirmed more recently by American studies. A British report estimated that the transmissibility of young people (up to the age of 16) is more than twice that of adults. Our neighbors across the Channel are all the more pessimistic, given that the variant that has emerged on their territory reproduces more easily in younger people than the historical strain, which we have also observed in France.

One of the difficulties is to have data on these age groups and to be able to distinguish between over-contagiosity (risk of transmitting the infection) and over-susceptibility (risk of developing the infection with equal exposure). As the youngest children are mostly paucisymptomatic (have few symptoms) or asymptomatic, they do not fit into the screening criteria (the French Pediatric Society has limited the indications for PCR screening for children under six), which restricts the possibility of uncovering and extinguishing chains of transmission through schools.

Another difficulty is determining the actual contact rate of children and adolescents. But one thing is certain: by definition, children have more contact than other age groups, because they are exposed to their peers. Should this overexposure be taken into account? Personally, I'm in favor of taking a rather parsimonious and conservative approach, in other words, not considering that children have a different profile to adults.

Unlike other teams, such as Vittoria Colizza's, we haven't worked specifically on schools. However, in our models, we don't make children more or less contagious or more or less susceptible than adults (on the other hand, we don't include them in the same way in hospitalization models, obviously, since severe forms only concern them very exceptionally).

As far as restrictions are concerned, it's clear that schools are places of contamination, whose closure represents a lever for curbing the epidemic. But it's also too essential to be considered as a primary target for restrictions.

"We tend to think that children are less likely to transmit the disease because they have asymptomatic forms.

Pascal Crépey, epidemiologist and biostatistician at École des Hautes Études en Santé Publique

What is clear is that it's complicated to track the dynamics of infection in schools, mainly because children don't have severe forms, and few have symptomatic forms. Unless you test them very regularly, whether they have symptoms or not, it's difficult to get a clear picture of the dynamics.

The question to be answered is: "Do children make a greater contribution than adults to the dynamics of the epidemic? Indeed, this is what would justify targeting them as a priority for restrictive measures.

However, the dynamics of influenza teach us that people without symptoms, even if they don't isolate themselves, are less contaminating than sick people. This is because they don't cough or sneeze, and therefore excrete fewer viral particles. In addition, although studies differ, their viral load is also lower, so they have, a priori, less virus to excrete.

On this basis, we tend to think that children are less likely to transmit the disease because they have asymptomatic forms. However, this is offset by the fact that the children will have a lot of contact with each other, and will certainly be less likely to respect physical distancing, the wearing of masks, hand-washing... However, since the reopening of schools, sanitary protocols have been put in place. While compliance with these protocols may vary from one school to another, they do further limit the risk of contamination at school.

As regards the installation of the variant of British origin, given that it is more transmissible than the historical strain at all ages, we can also expect an increase in contaminations among the youngest, but it will be of the same order as in the rest of the population. The fraction of contaminations attributable to children will not change.

In recent weeks, there has been a particularly strong screening effort in schools. Between week 10 and week 11, the number of tests performed increased by 40%. This increase has led to a rise in incidence, with more positive tests being found. However, the incidence among 10-14 year-olds remains lower than among adults. Infections do occur in children, but in more limited numbers than in adult populations.

"The question of keeping schools open is above all a question of benefit-risk balance".

Christèle Gras-Le Guen, Professor of Pediatrics, President of the French Pediatric Society

One year on, the role played by schools in the dynamics of the epidemic remains one of the points on which communication is most confused and difficult.

No one denies that schools are a potential source of contamination and could harbor infectious outbreaks: we've said that children are not very contagious, but we haven't said that they aren't contagious at all. However, we have observed that, in general, children are more likely to be contaminated by adults; the reverse is rarer. It's not impossible, but there's a difference between what's possible in theory and what we see in practice. Contamination occurs mainly within the family (meals and private gatherings).

Also read:
One year on, what do we know about the SARS-CoV-2 coronavirus infection in children?

We have published an article taking stock of the situation in France two months after the start of the school year in September. Not only is viral circulation among children and teenagers much lower than in adults, but very few classes have been closed and very few clusters investigated at school. In addition, figures from the French Ministry of Education show that very few teachers have been infected.

One point on which data is lacking is the question of the contagiousness of asymptomatic children, and their proportion. Up to now, children have been tested either because they had been in contact with a positive case, or because they were symptomatic, but no systematic tests have been carried out on non-symptomatic children. The advent of saliva testing in schools has made it possible to better describe the carriage of the virus in the absence of symptoms in children. The French Education Ministry's bulletin indicates that of the 200,404 tests carried out between March 15 and 22, only 0.49% were positive and therefore potentially contaminating, confirming that the contribution of asymptomatic children to the dynamics of the epidemic is marginal.

This is not to say that epidemic risk at school does not exist. However, it is minor compared to the expected benefit to children's health of keeping schools open, so that they can lead as normal a life as possible. This is all the more important in view of the fact that the mental health of the youngest children is particularly deteriorated by the health crisis. The question of keeping schools open is above all a question of benefit-risk balance, which must be adapted to the level of virus circulation and the age of the children, and is the last measure to be taken when all else has failed (strict confinement, optimized barrier measures, intensified vaccination of childcare professionals).

"Schools play a role in the dynamics of the epidemic".

- Dominique Costagliola, epidemiologist and biostatistician, Deputy Director of the Institut Pierre Louis d'Épidémiologie et de Santé Publique

I fully understand that people may decide that it's important, for many reasons, for children to continue going to school.

However, a policy that consists in declaring that "we must keep schools open" implies putting in place measures to properly manage the risk associated with this decision (ventilation, masks, distancing, screening...).

However, at the present time, given the very high circulation of the virus, I'm not sure this is possible. It's important to stress that the results published in the serious scientific literature have indeed shown that schools play a role in the dynamics of the epidemic.

On this subject, there is one point that is particularly worrying: how is it possible that the definition of a contact case in a school was not the same as that used by Santé Publique France and the Caisse Nationale d'Assurance Maladie everywhere else?

(editor's note: in schools, if a teacher is positive, his pupils are not considered at-risk contacts "because the teacher wears a mask" - if only one child is positive in a class, the other children are not at-risk contacts "because they are not very active in the virus transmission chain". A class was only closed once three cases had been confirmed. Things have recently changed in the reconfined departments: all classes will be closed "as soon as the first case is detected").

This definition of a contact case was used to imply that there was no problem in schools. Abroad, in Germany and England, schools were closed when severe measures were taken. In an article published in Le Monde on February 24, Mélanie Heard and François Bourdillon stressed the importance of recognizing the role of schools in the dynamics of the epidemic.

"If we want to keep schools open while bringing the epidemic more under control, we need to ensure that protocols are effective."

- Vittoria Colizza, Inserm Research Director at the Pierre Louis Institute for Epidemiology and Public Health

The susceptibility of children and their role in the dynamics of the epidemic were studied very early on in the pandemic.

To this day, the conclusions of this work have not changed: before the age of 20, individuals are less likely to be infected, and the contagiousness of young children is reduced.

There was probably an initial misunderstanding: some may have thought that children at school didn't catch the disease. This was not the case, as confirmed by the work of Arnaud Fontanet's team at the Pasteur Institute, which looked at the risk of infection among schoolchildren as a function of age. However, most cases went unnoticed because they were asymptomatic. Cases were only detected when a symptomatic case occurred, or after an investigation aimed at tracing contact cases in an infected household, for example.

Also read:
Covid-19: schools and contamination, what does science have to say?

The only thing that has changed in recent months is that the British variant is more easily transmitted and is the cause of more severe forms, which - in conditions of high incidence as in France today - means that more cases are now being detected in schools. What's more, in conditions where the virus circulates less widely elsewhere, as many places are closed and people telework, contamination in schools inevitably takes on greater importance.

One thing's for sure: if we want to keep schools open while bringing the epidemic under tighter control, we need to ensure that the protocols in place are effective.

Our latest work, carried out with Alain Barrat's team (CNRS), was aimed precisely at understanding whether regular screening in schools would reduce the number of cases. Based on contact data collected in a school with 250 pupils, we built a model for the spread of the epidemic within this school, located in a "reconfined" département. This model enabled us to compare the effectiveness of the protocol currently in place (class closure as soon as a case of Covid-19 is confirmed in a child) with various screening scenarios (PCR on nasopharyngeal swab, PCR on salivary swab, antigenic test on nasopharyngeal swab).

Closing the classroom according to the classic protocol reduces the number of cases in the school by 10-20% over one term. We compared this protocol with regular screening protocols: testing once every 2 weeks, once a week, twice a week, and every school day. We found that the key parameter was not test sensitivity, but screening frequency and adherence.

If only a quarter of students take part (low uptake, in the case of nasopharyngeal tests, which are unpleasant...), screening would have to be carried out almost every day to identify cases and thus further reduce the spread of the epidemic. If three quarters of pupils take part (in the case of salivary testing, which is less unpleasant), we can achieve the same level of impact reduction on the epidemic with just one test a week.

If we want to keep schools open as much as possible, we need to move towards this kind of protocol, which is much more effective than the current one. With the latter, class closure generally comes too late, and does not prevent the spread of the virus to other classes. In some Swiss cantons, tests are carried out on a weekly basis. In the UK, self-tests have been introduced, twice a week.

Lionel Cavicchioli, Head of Health, The Conversation

This article is republished from The Conversation under a Creative Commons license. Read theoriginal article.