Fox News has this article titled "Flu has killed 20K Americans including 136 children so far [as of March 2020] in this [2019-20] flu season, CDC says".

The CDC said that so far this season, 34 million people have gotten the flu, and about 20,000 have died.

Of those who died as of Feb. 29, 136 were children, the CDC said in its weekly flu report.

They link to the CDC's weekly flu report which, for the week ending February 29, 2020, which states (under "Key Points"):

CDC estimates that so far this season there have been at least 34 million flu illnesses, 350,000 hospitalizations and 20,000 deaths from flu.

Has the seasonal flu already killed 20,000 Americans during this season alone?

  • 16
    When does the season begin? – gerrit Mar 9 '20 at 8:33
  • 6
    Don't have a good source for the US but this is a very reasonable number. In Germany (approximately 1/4 of the population) there were 25100 influenca death in the 2017/2018 season, source Robert Koch institute, edoc.rki.de/bitstream/handle/176904/6253/… but this counted as a fairly bad year. – quarague Mar 9 '20 at 12:48
  • 1
    @gerrit: I don't know for the US, here in Germany the RKI (≈CDC) reports weekly on flu Oct - March and monthly over the summer. Start of the flu season is when acute respiratory disease numbers grow and there's lots of flu cases among them, this year (and the 2 years before) that was end of January/beginning of Feb. see influenza.rki.de/Wochenberichte/2019_2020/2020-09.pdf – cbeleites unhappy with SX Mar 9 '20 at 17:58
  • 11
    @Oddthinking Why did you copy the quote from the answer into the question? It makes the answer look unhelpful, since it just reiterates the question. – Barmar Mar 9 '20 at 19:17
  • 1
    @Barmar: Ah, I didn't actually look at the answer when I edited the question. I copied the quote because it was being directly and specifically referenced by the claimant. I consider it part of the claim - in fact, it IS the claim. – Oddthinking Mar 10 '20 at 1:22

Fox News links to the CDC source which says:

CDC estimates that so far this season there have been at least 34 million flu illnesses, 350,000 hospitalizations and 20,000 deaths from flu.

By "season" they mean the 2019/2020 season. For data on past seasons, see here.

Note the illnesses vs deaths here which gives a mortality rate of 0.05%.

The usual mortality rate of influenza (excluding the common cold) is 0.1%. Hospitalization rate for the flu is around 1%.

  • 11
    @Oddthinking While explicit mention of COVID-19 has been removed from the question, I think the relevance of the claim still depends on the connection. 'So far slightly below-average flu-deaths in 2019/2020' really only becomes noteworthy if it's combined with the (implied) claim that this shows that 'panic' of COVID-19 is exaggerated. I think the additional information about COVID-19 adds value insofar as it combats this false implied claim. – tim Mar 9 '20 at 12:46
  • 42
    Whether COIVID-19 is "worse" than the flu depends on a subjective choice of metrics. Whether the COVID-19 reactions (which vary immensely between countries and between individuals) are "panicked" depends on subjective opinions on what behaviours are justified given the risk. I don't think half-answers to half-baked speculations are helping much. It also doesn't help that the influenza's mortality is reasonably well-known, while the data on COVID-19 is still young and growing, so conclusions aren't very definitive yet. – Oddthinking Mar 9 '20 at 13:09
  • 35
    @Oddthinking By which metric is it not worse than flu? It has ~10-15x the hospitalization rate, ~20x time the death rate, the R0 is 1.5-2x higher up until now. The only metric in which COVID isn't worse is total number of cases... which is expected given that the virus started spreading only a couple months ago (not centuries like the flu) and the extraordinary measures that are being taken right now worldwide. In other words:the fact that COVID-19 is worse is not a subjective opinion, it's an established fact. – Bakuriu Mar 9 '20 at 19:55
  • 25
    @Bakuriu, we don't know the total infection rate for COVID-19. We know (approximately) how many people have gotten sick enough to see a doctor, but we don't know how many simply self-treated a mild infection, or are asymptomatic carriers. In contrast, the flu has been studied well enough that we can estimate those numbers. – Mark Mar 9 '20 at 21:08
  • 28
    @Bakuriu: I am not interested in arguing any of your figures. I am arguing the very definition of "better"/"worse" is a subjective one. Some people (including the OP, apparently) are using the "deaths so far" metric. If someone asks a question "Is X worse than Y", it needs to be changed to "Does X have a lower <metric used by claimant> than Y?" so we don't get caught in subjective arguments. – Oddthinking Mar 10 '20 at 1:19

The number of confirmed deaths due to influenza has been much less than 20,000 for the 2019-2020 flu season through March 5th 2020.

According the the US National Center for Health Statistics, the weekly data for influenza deaths so far this season (through the 8th week of 2020) is as follows:

Week       Deaths

 40 (2019)   16  
 41          16  
 42          18  
 43          30  
 44          31  
 45          29  
 46          39  
 47          49  
 48          65  
 49          94  
 50         106  
 51         122  
 52         180  
 1  (2020)  390  
 2          410  
 3          405  
 4          432  
 5          406  
 6          415  
 7          403  
 8          338  (data only 80% complete for this week)  

So about 4,100 so far.

NCHS collects death certificate data from state vital statistics offices for all deaths occurring in the United States. Pneumonia and influenza (P&I) deaths are identified based on ICD-10 multiple cause of death codes. NCHS surveillance data are aggregated by the week of death occurrence. To allow for collection of enough data to produce a stable P&I percentage, NCHS surveillance data are released one week after the week of death. The NCHS surveillance data are used to calculate the percent of all deaths occurring in a given week that had pneumonia and/or influenza listed as a cause of death.

However, as explained in this answer, modeling can be used to infer a higher number of deaths.

  • 4
    Influenza in elderly or weak patients can often progress to bacterial pneumonia and thence to death. It won't necessarily be recorded as a death caused by influenza. I don't find the figure of 20,000 implausible. Cold-bloodedly, the vast majority of such deaths were people whose time had run out. Covid19 kills younger and healthier people as well. – nigel222 Mar 10 '20 at 10:47
  • 1
    @nigel222 right, the 20,000 has a scientific basis, but it includes people who were never tested for influenza, and people for which, even using multiple different cause of death codes, influenza was not listed as one of the causes. If any of the multiple causes of death on the death certificate is influenza the death is included in the 4,100 number. – DavePhD Mar 10 '20 at 11:04
  • right. Even bacterial pneumonia may not be the immediate cause of death. It can lead to other more immediately fatal consequences such as heart attack, stroke, sepsis, failure of other or multiple organs. (Also not a few elderly patients have requested not to be resussitated should either of the first two happen). – nigel222 Mar 10 '20 at 11:27
  • If we venture in this direction and try to find the immediate cause of death (death probably being defined as the end of brain-function?) the virus itself will never be the thing which immediately does this. It will always be an indirect cause leading to the failure of central organs and the central nervous system, which we then define as death. - One other metric could be to try and estimate if the patient would have lived significantly longer without the presence of the influenza virus in his body. – Falco Mar 11 '20 at 13:08
  • @Falco the death certificates list multiple causes of deaths, and if any one of them is influenza it is included in the table in the answer (included in the 4,100 influenza deaths). – DavePhD Mar 11 '20 at 13:12

This Request For Correction (RFC) on the aspe.hhs.gov site gives detailed arguments that the CDC overestimates flu related deaths by uniquely bundling flu together with pneumonia, although there are other significant causes of pneumonia.

Its author identifies himself as "Kenneth Stoller, International Hyperbaric Medical Association".

Dr. Stoller begins as follows and continues with much detail which is worth reading and I am not sure of the appropriateness of copying it all here.

US data on influenza deaths are false and misleading. The Centers for Disease Control and Prevention (CDC) acknowledges a difference between flu death and flu associated death yet uses the terms interchangeably. Additionally, there are significant statistical incompatibilities between official estimates and national vital statistics data. Compounding these problems is a marketing of fear—a CDC communications strategy in which medical experts "predict dire outcomes" during flu seasons.

  • Welcome to Skeptics.SE! It's best to just quote the sections of the letter that are most important to answering the question. Copying the whole thing is redundant, especially since you have a link to an archived copy that shouldn't ever break. – F1Krazy Mar 12 '20 at 9:40
  • 1
    Phillip, I think you're on to some important information here, but total pneumonia deaths are roughly 60,000 so far this season (see spreadsheet linked in my answer), so the 20,000 isn't just from lumping the two together. – DavePhD Mar 12 '20 at 13:10
  • 3
    The CDC rejected Dr. Stoller's RFC. – David Hammen Mar 12 '20 at 13:20
  • 1
    @DavidHammen That is true. People can read the rejection letter and decide whether it addresses his critique, or whether it makes an appeal to authority. – Philip Mar 13 '20 at 4:51
  • 1
    I think @DavePhD's answer is excellent, as it provides definite data. If I could have, I would have merely attached the URL as a comment to his answer, but I do yet have the privilege. – Philip Mar 13 '20 at 6:53

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .