Question asked by André on 07/03/2020
You ask us the following question: “We hear very contradictory figures in the media regarding the number of deaths due to the flu: 8,000 per year for some, barely a few hundred at 8 pm in France 2. What is the reality?”
Thursday, on the television set, after several minutes devoted to Covid-19, Anne-Sophie Lapix launches a subject on the flu “Which would cause thousands of deaths every year in France”. Before taking the opposite foot (a little after 8 minutes): “In reality, it’s less, it’s even much less. The flu has killed 72 people since November 4. “ Later in the subject, the figure is repeated: “Nowadays [le 5 mars, ndlr], 72 people died from the flu. Last year at the same time, there were 194. “
Read also “Is seasonal flu more” deadly “than Covid-19?”
Similar figures are found elsewhere in the press. Like when South West written “With AFP” March 4: “The seasonal flu epidemic continues to ebb in mainland France, but has already caused 72 deaths.” A sentence taken from a Agence France-Presse dispatch of the same day. The day before in France-Soir, we read: “According to figures from Public Health France arrested on February 28, 55 people have died from the flu since November 4.”
These figures are surprising, because the same public agency referent in the matter was recording for the winter 2018-2019 “Approximately 8,100 deaths attributed to influenza”. This is the conclusion of an article written this summer on the subject.
Would the flu have become much less deadly, since it would have killed only 55 people between last fall and February 28 (and 72 if you push until March 4)? The answer is no. France 2, South West or France-Soir have simply read or misread the figures produced by Santé publique France. Confusion may be related to a change in the agency’s way of communicating – which can be explained by medical data that is not the same today as it was a year ago.
72 deaths… spent in intensive care
Each year, Public Health France, during the flu season, produces weekly notes detailing the progress of the disease. Take the Week 9 newsletter, released on March 4. The figure of 72 deaths is well mentioned. But this is by no means a total of deaths attributable to the disease.
This figure appears in the table “Description of severe influenza cases admitted to intensive care during the 2019-2020 season in mainland France (data included until week 9)”. And more precisely, in the row of the table: “Death among the cases admitted to intensive care”.
Translation offered to CheckNews by Public Health France: “This is the number of deaths observed among patients admitted for severe influenza in the 194 intensive care units participating in surveillance.” The complete public agency: “Resuscitation deaths represent only a small proportion of deaths from influenza.” Among other things because“A significant proportion of deaths directly attributable to influenza occur outside these services: hospital without intensive care, nursing home, home …” And to summarize: “It is wrong to say that the number of deaths that we publish each week in the” Surveillance of serious cases “section corresponds to all deaths attributable to influenza.”
Note that if the AFP writes at the start of its dispatch that “The seasonal flu epidemic continues to ebb in mainland France, but has already caused 72 deaths”, you have to wait a few lines to read: “A total of 72 deaths occurred among these intensive care patients.” A capital precision therefore, but absent in the recovery that the media sometimes made of this information.
No excess mortality
According to Public Health France, 72 people therefore died of the flu between November and the beginning of March, after being admitted to one of the resuscitation services participating in the surveillance of the epidemic. How many died in total? This data is not found anywhere in the March 4 bulletin.
In terms of communication, this is a big change from last year. In fact, in the bulletin produced in week 9 of the 2018-2019 season, we read from the start: “Approximately 5,700 deaths from influenza since the start of surveillance and up to week 7.”
Below, we also find the same table as that which persists in the weekly bulletins today: “Description of serious cases of influenza admitted to intensive care during the 2018-2019 season in mainland France (data included up to week 9) ” Where that number stands today at 72 in week 9, it was 194 for the same week.
Why was there a total death report last year and more today? To answer, you have to understand how the number of people killed by the flu is calculated each year. First, there must be an excess mortality for the population as a whole. If this is the case, this data from the civil status is cross-checked with information from the Public Health France network concerning the activity of the virus. A statistical model of the agency then allows it to know how much of the excess mortality is attributable to influenza.
However, if there was indeed an excess mortality during the 2018-2019 season, there is not this winter at this stage, the agency explains. This does not mean that there have been no deaths from the flu (nor have there been only 72). Simply that as it stands, the calculation method of Public Health France does not allow this value to be calculated for the whole country – since this “Influenza mortality model is applied when excess mortality is found”.