4 ms·

Apologies for the pedantry, but whether you're looking at national (US) or international numbers, COVID-19 has killed "only" about as many people as a flu seaso

by wk_end 6y ago

Apologies for the pedantry, but whether you're looking at national (US) or international numbers, COVID-19 has killed "only" about as many people as a flu season. ~300k have died from COVID-19 in the world, which matches the WHO's estimate of 290k to 650k per year for the flu, and ~80k in the US almost exactly matches American numbers for the flu a couple of years ago [2] [1] https://www.who.int/news-room/fact-sheets/detail/influenza-(seasonal) https://www.who.int/news-room/fact-sheets/detail/influenza-(... [2] https://www.statnews.com/2018/09/26/cdc-us-flu-deaths-winter/ https://www.statnews.com/2018/09/26/cdc-us-flu-deaths-winter... Moreover, comparing the flu season to "just 4 months" is disingenuous. That's...roughly as long or even longer than what most people consider the flu season. Of course, this is due to the lockdowns. I'm not trying to make the case that COVID-19 is "just a flu". we'd almost certainly be seeing much worse numbers if people weren't social distancing or whatever. And we're not finished yet.

DanBC 6y ago
You are not counting flu and covid using the same methods, which is causing you to either undercount covid deaths or overcount flu deaths. If you count covid and flu using the same method you see far more covid deaths. Here for covid you're using "died after testing positive for covid", and for flu you're using "died after testing positive for flu or with presumed flu". You can see these are different.
- wk_end 6y ago
  Sorry, I wasn't aware that I was doing that - I'm just using the official numbers I've got on hand. Can you provide any sources that provide an estimate that better lines up with the claim that COVID has "already killed 5x [more than] any flu season"? I know that different states and countries are reporting COVID deaths differently. I've seen "shutdown skeptics" claim that the reason Belgium and New York and Sweden's official numbers look so bad is because of this: supposedly they're using a methodology closer to the one you claim is being used to source the flu numbers, unlike everyone else. So I don't think my official numbers are purely what you claim they are, but probably a mix of different methodologies. Even if on balance that's producing an undercount, do we have evidence for something like an undercount by a factor of five? The former I'd be happy to accept, the latter is a bit of an extraordinary claim and calls for the evidence to match. Neither of my sources for the flu numbers say their number was derived in the way you're suggesting - the US one in particular says that it's an estimate using a statistical model rather than a count anything like what you're saying. Did I miss something in the sources?
  - anoncareer0212 6y ago
    The official numbers you have on hand are estimated using a ratio of deaths-to-hospitalizations [1] These official numbers are arrived at by _multiplying_ the number of flu death counts reported by various coefficients produced through complicated algorithms. The coronavirus numbers are _not_ the result of multiplying the a death count, they are the death count that is muliplied to get the flu numbers. You can find the observed flu death count for the last several years at https://www.cdc.gov/flu/about/burden/2017-2018.htm https://www.cdc.gov/flu/about/burden/2017-2018.htm, but you have to dig a little on the page for each year - the pages are for giving "Estimated Influenza Illnesses" and describing that years model, and in the citations you can find the count of observed. I'll come back and reply if I can make time to put together a spreadsheet with the observed counts per year with links. [1] https://www.cdc.gov/flu/about/burden/how-cdc-estimates.htm#References https://www.cdc.gov/flu/about/burden/how-cdc-estimates.htm#R... [2] https://blogs.scientificamerican.com/observations/comparing-covid-19-deaths-to-flu-deaths-is-like-comparing-apples-to-oranges/ https://blogs.scientificamerican.com/observations/comparing-... [3] https://aspe.hhs.gov/cdc-%E2%80%94-influenza-deaths-request-correction-rfc https://aspe.hhs.gov/cdc-%E2%80%94-influenza-deaths-request-...
  - DanBC 6y ago
    > the US one in particular says that it's an estimate using a statistical model rather than a count anything like what you're saying. To count deaths you can either count only the people who had a test and were positive, or you can count deaths where the disease is listed on the death certificate as a cause or contributory factor, or you can use complex statistical modelling and look at excess all cause mortality. (positives deaths is lower than death certificates, which is lower than the statistical modelling). Some countries further restrict the number given by testing. For example, the main number used in the UK during most of April was "was tested positive, died in hospital" -- they were not including care home deaths in that number. (They changed that in April so they now include care home deaths.) Here's a Flu Report from the UK for the 2015-2016 season: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/526405/Flu_Annual_Report_2015_2016.pdf https://assets.publishing.service.gov.uk/government/uploads/... If you look at tables six and seven they give all cause excess mortality for the flu seasons between 2006/07 through to 2015/16 (It's a bit annoying that they've done it by flu season and not by year).

by wk_end 6y ago

DanBC 6y ago

You are not counting flu and covid using the same methods, which is causing you to either undercount covid deaths or overcount flu deaths. If you count covid and flu using the same method you see far more covid deaths. Here for covid you're using "died after testing positive for covid", and for flu you're using "died after testing positive for flu or with presumed flu". You can see these are different.

wk_end 6y ago
Sorry, I wasn't aware that I was doing that - I'm just using the official numbers I've got on hand. Can you provide any sources that provide an estimate that better lines up with the claim that COVID has "already killed 5x [more than] any flu season"? I know that different states and countries are reporting COVID deaths differently. I've seen "shutdown skeptics" claim that the reason Belgium and New York and Sweden's official numbers look so bad is because of this: supposedly they're using a methodology closer to the one you claim is being used to source the flu numbers, unlike everyone else. So I don't think my official numbers are purely what you claim they are, but probably a mix of different methodologies. Even if on balance that's producing an undercount, do we have evidence for something like an undercount by a factor of five? The former I'd be happy to accept, the latter is a bit of an extraordinary claim and calls for the evidence to match. Neither of my sources for the flu numbers say their number was derived in the way you're suggesting - the US one in particular says that it's an estimate using a statistical model rather than a count anything like what you're saying. Did I miss something in the sources?
- anoncareer0212 6y ago
  The official numbers you have on hand are estimated using a ratio of deaths-to-hospitalizations [1] These official numbers are arrived at by _multiplying_ the number of flu death counts reported by various coefficients produced through complicated algorithms. The coronavirus numbers are _not_ the result of multiplying the a death count, they are the death count that is muliplied to get the flu numbers. You can find the observed flu death count for the last several years at https://www.cdc.gov/flu/about/burden/2017-2018.htm https://www.cdc.gov/flu/about/burden/2017-2018.htm, but you have to dig a little on the page for each year - the pages are for giving "Estimated Influenza Illnesses" and describing that years model, and in the citations you can find the count of observed. I'll come back and reply if I can make time to put together a spreadsheet with the observed counts per year with links. [1] https://www.cdc.gov/flu/about/burden/how-cdc-estimates.htm#References https://www.cdc.gov/flu/about/burden/how-cdc-estimates.htm#R... [2] https://blogs.scientificamerican.com/observations/comparing-covid-19-deaths-to-flu-deaths-is-like-comparing-apples-to-oranges/ https://blogs.scientificamerican.com/observations/comparing-... [3] https://aspe.hhs.gov/cdc-%E2%80%94-influenza-deaths-request-correction-rfc https://aspe.hhs.gov/cdc-%E2%80%94-influenza-deaths-request-...
- DanBC 6y ago
  > the US one in particular says that it's an estimate using a statistical model rather than a count anything like what you're saying. To count deaths you can either count only the people who had a test and were positive, or you can count deaths where the disease is listed on the death certificate as a cause or contributory factor, or you can use complex statistical modelling and look at excess all cause mortality. (positives deaths is lower than death certificates, which is lower than the statistical modelling). Some countries further restrict the number given by testing. For example, the main number used in the UK during most of April was "was tested positive, died in hospital" -- they were not including care home deaths in that number. (They changed that in April so they now include care home deaths.) Here's a Flu Report from the UK for the 2015-2016 season: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/526405/Flu_Annual_Report_2015_2016.pdf https://assets.publishing.service.gov.uk/government/uploads/... If you look at tables six and seven they give all cause excess mortality for the flu seasons between 2006/07 through to 2015/16 (It's a bit annoying that they've done it by flu season and not by year).