why are the indicators unstable?

Incidence rate, positivity, places in intensive care and other “R0”: given every evening, epidemic indicators have punctuated the last ten months of the life of the French and the government. However, on this New Year’s Eve, and while the fear of a third confinement is emerging, these scientific data appear particularly unstable.

More than 20,000 new positive cases were thus declared on December 25 against 3,000, only, the next day. A “yo-yo” of curves already observed since the arrival of SARS-CoV-2, but never as frantic as during this holiday season.

→ READ. Covid-19: towards local reconfigurations after the holidays?

However, these figures continue to dictate the action of the executive and could justify further restrictions. The Minister of Health, Olivier Véran, recalled them on December 27, in the Sunday newspaper, to alert on the situation: “The virus is still circulating too much: 15,000 contaminations detected per day on average, whereas we had fallen to 11,000… The objective of 5,000 is moving away. “

The pitfall of Sundays and public holidays

Difficult, precisely, to be clear on the exact level of new positive cases for Covid-19. Less than 3,000 were identified on December 28 by Public Health France (SPF). That is three times less than the day before or seven times less than on December 24. However, there is no question of seeing a sudden drop in contamination in the country.

Like every Sunday since the arrival of the virus, public holidays see a large part of laboratories and pharmacies lower their curtain. The closure of these screening places naturally results, the next day, in a drastic drop in the number of tests – and positive cases – reported by the health authorities.

Likewise, daily hospital data should be handled with caution. The new admissions recorded on Sundays are each time much lower than the rest of the week. 599 additional hospitalizations were thus counted on December 27, against 1,217 the next day or 1,454 three days earlier. Variances which could be explained by the lack of staff on weekends in hospitals to collect and report the data, which are corrected at the beginning of the week.

Testing rush before Christmas

To offer a faithful overview of the epidemic reality, daily data is therefore not sufficient. Not even the figures from week to week, in this particular context of the end of the year and the holidays. Beware of biased readings.

Thus, from December 14 to 20, 98,280 new carriers of SARS-CoV-2 were recorded, an increase of 23% compared to the previous seven days. The same period was however marked by an even greater increase in the number of people tested (+ 74%). A rush on the swabs of those who wanted to protect themselves before Christmas Eve or took part in the massive screening operations organized by certain communities.

« People without any symptoms started to be tested, to reassure themselves, confirms Jean-Stéphane Dhersin, Scientific Deputy Director of the National Institute of Mathematical Sciences and their Interactions of the CNRS (Insmi-CNRS). Obviously, this induces a larger number of positive cases but, being mostly negative, they lowered the positivity rate. [passé de 6,1 % à 4,3 % entre les semaines 50 et 51]. This adds difficulty in estimating the parameters. ”

Stagnation in hospitals

This specialist in epidemic modeling invites us to wait for the “Real figures, those of hospitalizations, resuscitations and deaths”. “It is quite difficult to know if this rise [des personnes positives] is only due to the increase in the number of tests, insists the mathematician. The real indicator will be that of hospitalizations, but with a fifteen day lag. “

→ ANALYSIS. Covid-19: in Europe, first injections and first reconfinements

We will therefore have to wait until the end of the first week of 2021 to conclude that there is a possible epidemic resumption in hospitals, and to verify the concrete effect of the end-of-year celebrations on the circulation of the virus. For ten days, the number of people hospitalized (around 25,000) and admitted to critical care (2,700) has stagnated at levels recorded on the eve of the second confinement, at the end of October. Two markers of the famous « plateau » which makes the government fear so much a new rebound.


Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.