January 26, 2022

Omicron variant: the Institut Pasteur scenarios for the peak hospital

SCIENCE – Records are still broken daily, but the peak in coronavirus cases is likely to be in a few days. In this case, the peak in hospitalizations will take place by the end of January, but its height will depend on several parameters. Here are the conclusions of the latest analysis by the Institut Pasteur on the impact of the Omicron variant on the Covid-19 epidemic.

Finalized on 7, but published this Wednesday, January 12, this work is testing several scenarios to try to glimpse what the state of the hospital system might look like in the coming weeks, depending on various hypotheses.

There are still many uncertainties, as the characteristics of the Omicron variant are still unclear. In the worst-case scenario imagined by Pasteur’s teams, the peak of the first wave of 2020 could be exceeded, both in terms of hospitalizations and intensive care admissions. Fortunately, this is not the most likely scenario.

In the most optimistic scenario, the rise in hospital indicators will be slight and the peak much lower than that of the first wave. But this scenario is not the most likely either.

Projections that depend on us and vaccines

The purpose of these theoretical models is not to predict the future, but to help understand how changing certain elements of the equation changes the game. “For example, if the French reduce their contacts by 20% in January, this could reduce the size of the peak in hospitalizations by 50%,” the authors note.

To arrive at this result, the researchers looked at the scientific literature about Omicron to estimate its contagiousness, its virulence (the risk of severe form, lower than for Delta), its resistance to vaccines. With these parameters in hand, they imagined three main scenarios:

  1. A severity twice lower than that of the first strain of the coronavirus but a high transmissibility (the most likely scenario for the researchers in view of the latest data)
  2. A severity identical to the original strain (but therefore lower than Delta) and an intermediate transmissibility
  3. Same severity as the original strain and high transmissibility.

Then, they tested in their model how different parameters could modify the peak of hospital admissions, the number of beds in conventional hospitalization and in critical care.

In the graph below, the researchers tested three different projections based on these three scenarios. In the first, the red curve, the French do not decrease their number of contacts compared to December (they do not see less people). In the second, in green, this number is reduced by 10% and in the third, in blue, by 20%. The dashed horizontal line represents the peak of the first wave of 2020.

In the best-case scenario, the peak in hospital admissions would be 2,500, which seems very optimistic when we see that today, when the peak of cases is not yet here and there is a delay between cases and hospitalizations, we are already at nearly 2,200 daily admissions.

But we should not believe that this necessarily means that we have not reduced enough our contacts at risk. Again, these projections are theoretical and depend on many parameters.

Many unknowns

The researchers thus launched their model with the same projections and scenarios, but by changing an important variable: vaccine effectiveness in blocking infection. Indeed, according to some British data, the vaccine would be less effective than what the Pasteur Institute had anticipated.

A two-dose regimen of more than six months would protect 10% against 25% of infection. As a reminder, we would go from 70% to 40%. The protection against hospitalization, which is very high, remains unchanged.

Taking into account this more pessimistic vaccine efficacy, the peaks are much higher in general, even if it is still possible to remain under the impact of the first wave thanks to a small reduction in our contacts.

Another important variable: the number of people vaccinated and having recalled. Thus, the Institut Pasteur is based on the rate at the end of December of 800,000 injections per day, while we are more around 560,000 since the end of the school holidays. Fewer people with a recall means more risk of being infected and therefore of spreading Omicron and therefore, ultimately, more hospitalizations.

There are also variables that could change for the better. For the moment, the model of Pasteur’s teams estimates that a hospitalized person is at as much risk of going into intensive care, whether they are vaccinated or not. However, “it is however probable that the hospitalizations are less serious for the vaccinated persons, this could lead to overestimating the number of beds necessary in critical care”.

Finally, the Institut Pasteur also calculated the impact of a reduction in hospital stay time. By halving the average length of stay, this reduces the need for conventional hospital beds by 34%. “This requires the development of primary care structures or home hospitalizations for the reception of patients upstream or downstream of hospitalizations”, report the authors of the study.

The ideal would obviously be not to have to come to this to contain a peak of Omicron too important.

See also on The HuffPost: how to properly use a self-test?