CORONAVIRUS – It is now an exponential rise. The weekly figures communicated Thursday evening August 13 by the General Directorate of Health (DGS) on the situation of Covid-19 in France leave no room for doubt: with + 41% for the incidence rate and + 42% for the positivity rate of tests in one week, France is faced with a resumption of the epidemic of coronavirus.
There are now 33 departments with more than 10 patients per 100,000 inhabitants: the Provence-Alpe-Côtes d’Azur region is particularly under surveillance. The only good news comes from the Mayenne, the epidemic focus of the last few weeks, which saw its incidence rate drop from 146 to 83, in “very clear reduction” indicate the authorities.
The number of hospitalizations of patients with Covid is also on the rise, for the third week in a row: 782 new admissions against 604 the previous week.
Paris and the Bouches-du-Rhône in “alert threshold”
To see the situation a little more clearly, The HuffPost a mis au point a map of France of the circulation of the coronavirus, updated automatically. This shows this new situation well. When we first published it on July 23, only four departments had exceeded a vigilance threshold in metropolitan France (in light green). There are now more than thirty.
Paris and the Bouches-du-Rhones are now in “alert threshold” (in yellow on our map) and this Friday passed into an active circulation zone of the virus, according to a decree published this Friday August 14 in the Official Journal. The prefects of the two departments will therefore be able to take more restrictive measures. Depending on the situation, they could for example “decide to close restaurants, markets or limit gatherings even more.
We first chose to base ourselves on two main elements: the incidence rate (how many cases per 100,000 inhabitants) and the test positivity rate. Each department is colored according to the evolution of these indicators. The first card (global index button) gives a general view of the state of the situation. The second (trends button) allows you to see the evolution of the indicators over time.
For technical reasons, the overseas territories are not visible, but are accessible in the search engine at the top left (find below our explanations and our methodology). This map updates automatically as soon as Public Health France updates the data posted on data.gouv.fr (almost daily).
How is this map made?
To follow the evolution of the epidemic, the government and the health authorities are based on a large number of indicators, more or less reliable. Others should be added to this arsenal shortly, such as wastewater analysis.
Four indicators are particularly monitored and have “vigilance” and “alert” thresholds. When a threshold is crossed, “an in-depth risk analysis is launched” in the geographical area concerned, precise the Ministry of Health. Here are these 4 indicators:
- Incidence rate, or the number of positive PCR tests per 100,000 inhabitants per week. The vigilance threshold is at 10, the alert threshold at 50.
- Test positivity rate, the vigilance threshold is at 5%, the alert threshold at 10%.
- The share of intensive care beds occupied in hospitals, the vigilance threshold is 40%, the alert threshold is 60%
- The effective R, the reproduction rate of the virus. If it is less than 1, the epidemic decreases, if not, it increases more or less quickly. The vigilance threshold is at 1, the alert threshold at 1.5.
To produce the map above, we have chosen to base ourselves mainly on the incidence and positivity rate (smoothed on the average of the last 7 days). In a context of risk of a resumption of an epidemic, the proportion of intensive care beds does not seem the most appropriate. Theoretically, the other indicators should indeed rise well before that.
As for the effective R, if it can be very useful, its method of calculation is still not public. In addition, the data made available by the government and Public Health France are not very clear. The updates are partial, irregular and are limited to regions (no differentiation by department). We have indicated the actual R in the detailed information for each department, but have decided not to color the map according to this index until its updates are calibrated to the same format as that of the other two indicators.
Two complementary indicators
The incidence and positive rates are interesting to read together. The first allows to know if, in an area with a given population, there are many cases of coronavirus. But alone, it can be misleading. Because we know that the virus circulates at low noise. And that sometimes when the conditions are right, large clusters may appear.
This is why this incidence rate, this presence of the coronavirus, must be compared with other indicators. The positivity rate, in particular, is useful for this. If there are a lot of cases in a territory (incidence rate), but this is only due to highly developed screening, the positivity rate will be low. Conversely, if it increases, it means that most of the people tested are positive, but especially that the infected people who are not tested, who fall through the cracks, are even more numerous.
If they are not perfect, these two cumulative indicators therefore already provide a good idea of the situation in a given department.
We have also developed a second map, visible by clicking on the “trends” button. This does not make it possible to visualize the thresholds of these indicators, but their evolution over time. Because what is important, in an epidemic, is the trend of the curve: if it is on the rise, the risk is that it will be exponential, that it rises more and more quickly. Conversely, if the incidence or positivity rate is high, but the clusters are well controlled, these indicators will drop.
This second version of the map therefore checks the average evolution over the last 7 days of these two indicators to get an idea of the departments where the epidemic seems to be progressing or declining.
See also on The HuffPost: Ranking of masks according to their effectiveness against the coronavirus