Why projections on the Covid waves are becoming more and more complicated

SCIENCE – Started at the beginning of June, we can say this Tuesday, July 26, that the 7th wave of Covid-19 in France reached its peak. On the side of positive cases like hospitalizations, the indicators are down, as you can see in the graph below.

Good news (if the trend continues), because this peak associated with the BA.5 variant has just equaled the one associated with BA.2, far from the first “Omicron wave”, caused by BA .1. Good news too, because although the situation in Portugal and South Africa made it possible not to get too alarmed, it was the first wave that the Scientific Council (and therefore the government) had no models to lean on.

And for the time being, the Institut Pasteur also does not have a model for the future waves that will undoubtedly come as soon as the summer is over and our immunity will have slowly declined.

This did not prevent the Scientific Council, in its last opinion issued on 19 July and detailed at a press conference, from working on three scenarios for the autumn. But these lanes are very generic: a return of the existing variants, a descendant variant of Omicron, or worse, a very different and possibly much more dangerous variant.

Models become too complex

But why does the Scientific Council no longer have projections that make it possible to predict? “This BA.5 wave was the first where we did not have a model because over the last two years we started with simple models which gradually became more complex to integrate the impact of the epidemic of variants, vaccines, as well as a decrease in immunity,” explains Simon Cauchemez, modeler at the Institut Pasteur and member of the scientific council. “These overly complex models are today uncertain”.

To make projections of the curves for Covid-19 (more details in this 2020 interview with Simon Cauchemez), we formulate hypotheses about the virus (its contagiousness, duration of infection, severity, etc.) and about the target population (the number of infectious people , the number of exposed contacts, age, etc.).

At the start of the epidemic, things were (unfortunately) simple: Almost anyone could catch Covid-19. Even after the first wave, which only hit 5% of the population. There were only two ways to reach the top. Either let the virus spread until there aren’t enough people to be infected (leaving tens or even hundreds of thousands of people to die). Either take measures that limit our contacts at risk to break the chains of transmission of the coronavirus.

But since then things have evolved in many ways. First, thanks to vaccines, which have given us very significant protection against severe forms and, in a lighter and more fleeting way, against infection. There were also the variants that had to be integrated into the models. Did this new set of mutations make the virus more contagious? Less virulent? Can you escape the vaccine? For a previous infection?

“We know little about cross-immunity between variants”

All these parameters made the projections more complex, but the models still saw things clearly. “Until then, you could simplify by putting people in boxes. People who are vaccinated with one dose, two doses, a booster, those who are infected with natural immunity”, explains Samuel Alizon, director of research at the CNRS, specialist in modeling infectious diseases. “But the Omicron waves blew up the categories.”

With the arrival of the highly contagious Omicron variant, most Western countries, widely vaccinated, tired of repeated containments and unable to develop non-coercive containment measures, chose to let the epidemic slip away. By doing this, we accepted a very large wave of hospitalization cases, but with a much lower toll than for the previous variants on an unvaccinated population. There was also the vague and ludicrous hope that this wave would be the last, causing “herd immunity”, preventing the coronavirus from circulating.

But reality has again very quickly caught up with the flimsy hopes. This natural immunity, as we already knew, does not last forever. It wanes and disappears over time (although against severe forms it seems to stabilize after three doses or three infections).

And that’s a big part of the problem. “While vaccine immunity is easy to control and monitor, natural immunity is less well known,” explains Samuel Alizon. Especially with the multiplication of variants and situations. In which box should a vaccinated person, contaminated 3 months later, who then had a recall in January? How comparable is his immunity to someone infected in 2020, vaccinated 2 times in 2021 and then re-infected with BA.1 in January 2022? Or at BA.2 in March?

“We know a little about cross-immunity between variants, for example we have seen that BA.5 can bypass part of the immunity caused by an infection with BA.1”, illustrates Samuel Alizon. All these boxes therefore become very difficult to control, so that an epidemiological model can offer precise projections without risking going completely astray.

Simplify without distorting

However, it is always necessary to anticipate as much as possible. “The pandemic is not over. We are facing a virus that has a genetic evolution that is difficult to predict”, warned Jean-François Delfraissy, the president of the Scientific Council, as a preamble to present his latest statement.

But can we even adapt the calculations to this new situation? “Today’s models are too complex and therefore unstable. A compromise must be found with more parsimonious models taking into account these different immunity profiles. It is a work in progress”, explains Simon Cauchemez.

In a pre-published paper on June 15, Samuel Alizon and two colleagues tested a new concept in an attempt to account for the decline in immunity. “The idea is to include in the model how long individuals have been in a certain state, for example how long since their last dose of vaccine,” he explains. “One of the results is that even in the absence of a new variant, large annual waves associated with winter and the (limited) decrease in immunity can be observed”.

Surprisingly, the scenarios in which the entire population is vaccinated at the same time each autumn lead to a more pronounced peak than if a booster is offered only to the elderly and infirm (although the more vaccinated, the fewer deaths there are). The reason put forward by the researchers: by vaccinating everyone at the same time, the level of immunity is synchronized. It is clear that we suddenly have many people who become susceptible to infection again. Another lesson from the study, the researcher notes: “In addition, so-called non-pharmaceutical interventions (improving air quality, wearing a mask, etc.) can have an effectiveness comparable to annual vaccination campaigns. Ultimately, the best effectiveness is achieved by combining these interventions and vaccine boosters.”

Obviously, this type of general projection has limitations. “The further we project ourselves into the long term, the more qualitative the model becomes”, clarifies Samuel Alizon. What the health authorities want, however, are “quantitative” projections. To put it simply, let’s say a qualitative model tries to imagine the general trend of the curve for Covid-19 over the long term under various assumptions. The quantitative model will try to predict the number of people infected or hospitalized. “But as soon as we go beyond the month, these quantitative models become delicate in the face of the many unknowns, and we still have to explore different scenarios.”

To conclude, we must remember that we are obviously not powerless to monitor this pandemic. “We will have to be vigilant about the next eruptions, because it is very difficult to say the dates and the size of the peaks. Today we observe what is happening to our neighbors and it is very instructive”, recalled Arnaud Fontanet, epidemiologist and member of the Scientific Council, during the press conference. However, the future dominant variant must not appear in France. “If we are on the front line, it will be difficult, we will have to remember the possibility of a slightly more disruptive emergence”.

Also look at HuffPost: Why vaccination of people over 60 is not superfluous

Leave a Comment