It seems every journalist has his favorite expert that they can quote to whip up a CFR or IFR number for influenza. (For example, a question here [which no longer has this information] said that the IFR for influenza is 0.1%, although I think the unclear passages in the sources cited e.g. are actually talking about CFR. A Bloomberg piece which is more clear/explicit that it is talking about IFR puts it at 0.04%, but again the source is some expert interviewed.)

It seems substantially harder to find some meta-analysis giving such IFR figures for influenza.

What I could find in the peer-reviewed literature is that the asymptomatic rate for influenza isn't a slam dunk:

Most point estimates from studies of outbreak investigations fell in the range 4%–28% with low heterogeneity (I2=0%) with a pooled mean of 16% (95% CI: 13%, 19%). Estimates from the studies conducted across epidemics without adjustment were very heterogeneous (point estimates 0%–100%; I2=97%), while estimates from studies that adjusted for background illnesses were more consistent with point estimates in the range 65%–85% and moderate heterogeneity (I2=58%). Variation in estimates could be partially explained by differences in study design and analysis, and inclusion of mild symptomatic illnesses as asymptomatic in some studies.

So it doesn't look like one can resort to a generally accepted asymptomatic ratio and simply multiply any random CFR from influenza studies with that.

Not that the CFR for influenza seems very easy to pin down either, e.g. a meta-analysis on 77 studies on the H1N1/pdm09 CFR found a large range. Oddly, this paper's abstract doesn't report the more usual meta-analytic attempts at some central value with confidence intervals.

We identified very substantial heterogeneity in published estimates, ranging from less than 1 to more than 10,000 deaths per 100,000 cases or infections. The choice of case definition in the denominator accounted for substantial heterogeneity, with the higher estimates based on laboratory-confirmed cases (point estimates = 0-13,500 per 100,000 cases) compared with symptomatic cases (point estimates = 0-1,200 per 100,000 cases) or infections (point estimates = 1-10 per 100,000 infections). Risk based on symptomatic cases increased substantially with age.

So yes, they estimated the IFR for H1N1/09 to just 0.001% to 0.01%. But supposedly the seasonal influenza, H3N2 in particular, is more deadly than H1N1 (but even that relative proportion might vary with the age group.)

But perhaps there is less disagreement on the CFR for seasonal influenza.

So, are there such IFR (not CFR) estimates for seasonal influenza? In particular, I'm most interested in global estimates rather than focused on one country and in meta-analytical results rather than single (outbreak/season) studies, but I'll relax my requirements if they are too strict in producing any answer.

  • in an unimmunized population? May 10, 2020 at 0:03
  • @GrahamChiu: in whatever they managed to study... (if there's a notable difference [and enough studies in each group], I'd expect a meta-analysis to suss it by metaregression) May 10, 2020 at 0:05
  • It's a bit pointless if comparing to covid-19 which is an unimmunized population May 10, 2020 at 0:09
  • @GrahamChiu: I'm not asking for any comparisons with Covid-19. May 10, 2020 at 0:10
  • 1
    Yes, but this question is motivated by the other one for covid-19 vs flu May 10, 2020 at 0:12

1 Answer 1


“ Peer-reviewed IFR estimates for (seasonal) influenza?”

This isn’t really an answerable question as is because influenza isn’t an illness. It’s a group of illnesses caused by a pretty wide variety of infectious agents. For starters: Influenza A&B last year in the US

and it will vary from year to year:

annual flu chart

H1N1? H3N2? Yamagata? Victoria? Not equally lethal. Stats per strain

Aside: Although nominally it’s not a question you’re asking, I would suggest you look at the number of deaths,total, in places where the SARS-CoV-2 virus has spread the most, like New York City. See row below, but NOTE: NYC deaths are around 26,000, last I checked, per NYTimes. So nearly .2% of NYC is dead due to COVID.

Here’s a graphic from Twitter by @Thomas Kirch chart

I mentioned it here instead of in a comment so that I can post the image; forgive the rule bending.

  • Can you please cite where you obtained the data and graphs you have provided here? Also, a heads up as providing information as images with no information in text extrapolating and explaining the results to 1) those who cannot see the image due to blindness, and 2) those who may not understand the information in the image is frowned upon. Jan 21, 2021 at 19:32
  • 1
    Also, images aren't searchable. It's unlikely that a web search (or site search) will find your answer even if someone searches for the words that appear in your post. And without links to the source of the images, readers can't fall back on going to the original.
    – Carey Gregory
    Jan 21, 2021 at 19:42
  • You have a (very) valid point here with influenza not being a single pathogen, but some summarization of the findings (with links/citations) would be substantially better than the large number of screenshots. (A bit ironically, the last table disproves your point a bit, as they do compute an average CFR.) A more useful measure to prove your point is some measure of the dispersion of the results. Jan 23, 2021 at 3:11
  • Chris, Carey: Yes, a problem for the blind, sadly. Disabled myself. Improvements welcome. These were screen shots I'd taken prior to seeing the question. There is text between each one explaining their relevance, and each is at least partly self-documenting as to source except the last one, for which I provide the source. They are all from the MMWR part the site noted (cdc.gov) or an archive thereof, which should be easy to find on the site, with a google searches with terms such as site:cdc.gov FluView MMWR "[quoted substring from an image]". Jan 28, 2021 at 2:40
  • Sorry; not ideal/formal but certainly makes 'em all readily findable. Jan 28, 2021 at 2:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.