4 ms·

My understanding is that it’s the other way around — both cases and deaths are being undercounted substantially. And testing capacity has been intentionally kne

My understanding is that it’s the other way around — both cases and deaths are being undercounted substantially. And testing capacity has been intentionally kneecapped by the govt (if you test fewer, you’ll find fewer cases, which looks good). All told, COVID looks to be about 10x deadlier than the flu. Which is still low, but not trivial given how quickly it spreads.

thu2111 6y ago
I don't think your understanding is up to date. 10x deadlier than the flu would require it to have an IFR of 1%. Even the US CDC, which has been over-exaggerating the dangers from the start, peg it at much less than that, more like 0.3% and many studies are putting it even lower, like between 0.08% (low end, Denmark) and 0.36% (Germany, Heisenberg). But those IFRs are also calculated using the standard definition of a COVID death - someone who died, and was testing positive. If you look at Italy, early on they realised there was a conflation going on, and so they did proper studies of some sample of COVID deaths to see why they really died. They concluded only 4% of the patients had no other medical conditions, and I recall reading (but don't have a link right now) that they concluded about 12% had really died of COVID. The rest all just happened to be infected when they died of something else, which would make sense given the first stat. So you can cut a digit off most of the reported death rates. All suggestions of undercounting are based on the assumption that all excess death is caused by COVID, and nothing else, even though in some countries e.g. the UK many excess death certificates don't even mention COVID, and in other countries like Germany there basically isn't any excess death: https://www.destatis.de/EN/Themes/Cross-Section/Corona/Society/population_death.html https://www.destatis.de/EN/Themes/Cross-Section/Corona/Socie...
- malyk 6y ago
  If you look at the CDC data on Excess Deaths[1] then you can see that COVID and the response to COVID has been responsible for a significant number of excess deaths in every week since the week ending on 3/28. So, maybe not 10x deadlier, but certainly way deadlier in total count of deaths. 1 - https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm
  - thu2111 6y ago
    COVID and the response to it - maybe yes. But remember that many countries have seen no excess death at all, their graphs are totally flat. Just look at EuroMOMO. People have tried to figure out why. It's supposedly the same virus everywhere. It's not related to hospital overload or healthcare availability because nowhere ran out of that, despite the initial scares. It might be care home related, as that's where 80% of the deaths are (reported to be). It might be related to decisions to flush hospitals to make way for a modelled surge that never happened. But then that'd fall squarely into the response bucket. The US data is hard to interpret because it only goes back to 2017. I can't seem to find longer term data. Do you know where to find it? In the UK - one of the 'worst hit' in the world supposedly (but those stats were found to be wrong) - to find similar excess deaths you have to go back to 1998/1999/2000. Is this bad? Well, "once in 20 year excess death spike" may seem bad. But events that happen once every 20 years aren't especially rare. Back then nobody panicked. Nobody talks about the terrible winters of 1999 when they lost so many loved ones. So it's really important to be able to put these excess death stats in perspective, otherwise you just see a bump in a graph and are told "this is really bad".
- mftrhu 6y ago
  > If you look at Italy, early on they realised there was a conflation going on, and so they did proper studies of some sample of COVID deaths to see why they really died. They concluded only 4% of the patients had no other medical conditions, and I recall reading (but don't have a link right now) that they concluded about 12% had really died of COVID. The rest all just happened to be infected when they died of something else, which would make sense given the first stat. So you can cut a digit off most of the reported death rates. If you look at Italy, the recent ISTAT-ISS report (2020-07-16) showed the opposite: COVID-19 has been deemed directly responsible for the death of 89% of those who died testing positive for coronavirus ("morti per"), with the remaining 11% of deaths being caused by some other disease or pre-existing condition ("morti con"). Even excluding comorbidities such as diabetes and hypertension, which are very common, 18% to 28% of the deaths they analyzed were still directly caused by COVID. I'm not sure this can be brought down to 12% without being exceedingly selective (e.g., people aged 0-30 without hypertension, diabetes, respiratory problems... - I know I'm 27 and I wouldn't qualify). > - Sono state analizzate le informazioni riportate dai medici in 4.942 schede di morte, di soggetti diagnosticati microbiologicamente con test positivo al SARS-CoV-2 (il 15,6% del totale dei decessi notificati al Sistema di Sorveglianza Integrata ISS fino al 25 maggio). Nelle schede di morte sono certificate, oltre a COVID-19, quelle condizioni e malattie che hanno avuto un ruolo nel determinare il decesso. > - COVID-19 è la causa direttamente responsabile della morte nell’89% dei decessi di persone positive al test SARS-CoV-2, mentre per il restante 11% le cause di decesso sono le malattie cardiovascolari (4,6%), i tumori (2,4%), le malattie del sistema respiratorio (1%), il diabete (0,6%), le demenze e le malattie dell’apparato digerente (rispettivamente 0,6% e 0,5%). > - La quota di deceduti in cui COVID-19 è la causa direttamente responsabile della morte varia in base all’età, raggiungendo il valore massimo del 92% nella classe 60-69 anni e il minimo (82%) nelle persone di età inferiore ai 50 anni. > - COVID-19 è una malattia che può rivelarsi fatale anche in assenza di concause. Non ci sono infatti concause di morte preesistenti a COVID-19 nel 28,2% dei decessi analizzati, percentuale simile nei due sessi e nelle diverse classi di età. Solo nella classe di età 0-49 anni la percentuale di decessi senza concause è più bassa, pari al 18%. Televideo source: http://www.televideo.rai.it/televideo/pub/view.jsp?id=888&p=101 http://www.televideo.rai.it/televideo/pub/view.jsp?id=888&p=... ANSA source: https://www.ansa.it/canale_saluteebenessere/notizie/sanita/2020/07/16/studio-istat-iss-covid-causa-diretta-di-morte-per-l89-di-positivi_dda57022-331d-4142-a163-6966e3e87916.html https://www.ansa.it/canale_saluteebenessere/notizie/sanita/2... The ISS-ISTAT report itself: https://www.istat.it/it/files/2020/07/Report_ISS_Istat_Cause-di-morte-Covid.pdf https://www.istat.it/it/files/2020/07/Report_ISS_Istat_Cause...

thu2111 6y ago

I don't think your understanding is up to date. 10x deadlier than the flu would require it to have an IFR of 1%. Even the US CDC, which has been over-exaggerating the dangers from the start, peg it at much less than that, more like 0.3% and many studies are putting it even lower, like between 0.08% (low end, Denmark) and 0.36% (Germany, Heisenberg). But those IFRs are also calculated using the standard definition of a COVID death - someone who died, and was testing positive. If you look at Italy, early on they realised there was a conflation going on, and so they did proper studies of some sample of COVID deaths to see why they really died. They concluded only 4% of the patients had no other medical conditions, and I recall reading (but don't have a link right now) that they concluded about 12% had really died of COVID. The rest all just happened to be infected when they died of something else, which would make sense given the first stat. So you can cut a digit off most of the reported death rates. All suggestions of undercounting are based on the assumption that all excess death is caused by COVID, and nothing else, even though in some countries e.g. the UK many excess death certificates don't even mention COVID, and in other countries like Germany there basically isn't any excess death: https://www.destatis.de/EN/Themes/Cross-Section/Corona/Society/population_death.html https://www.destatis.de/EN/Themes/Cross-Section/Corona/Socie...

malyk 6y ago
If you look at the CDC data on Excess Deaths[1] then you can see that COVID and the response to COVID has been responsible for a significant number of excess deaths in every week since the week ending on 3/28. So, maybe not 10x deadlier, but certainly way deadlier in total count of deaths. 1 - https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm
- thu2111 6y ago
  COVID and the response to it - maybe yes. But remember that many countries have seen no excess death at all, their graphs are totally flat. Just look at EuroMOMO. People have tried to figure out why. It's supposedly the same virus everywhere. It's not related to hospital overload or healthcare availability because nowhere ran out of that, despite the initial scares. It might be care home related, as that's where 80% of the deaths are (reported to be). It might be related to decisions to flush hospitals to make way for a modelled surge that never happened. But then that'd fall squarely into the response bucket. The US data is hard to interpret because it only goes back to 2017. I can't seem to find longer term data. Do you know where to find it? In the UK - one of the 'worst hit' in the world supposedly (but those stats were found to be wrong) - to find similar excess deaths you have to go back to 1998/1999/2000. Is this bad? Well, "once in 20 year excess death spike" may seem bad. But events that happen once every 20 years aren't especially rare. Back then nobody panicked. Nobody talks about the terrible winters of 1999 when they lost so many loved ones. So it's really important to be able to put these excess death stats in perspective, otherwise you just see a bump in a graph and are told "this is really bad".
mftrhu 6y ago
> If you look at Italy, early on they realised there was a conflation going on, and so they did proper studies of some sample of COVID deaths to see why they really died. They concluded only 4% of the patients had no other medical conditions, and I recall reading (but don't have a link right now) that they concluded about 12% had really died of COVID. The rest all just happened to be infected when they died of something else, which would make sense given the first stat. So you can cut a digit off most of the reported death rates. If you look at Italy, the recent ISTAT-ISS report (2020-07-16) showed the opposite: COVID-19 has been deemed directly responsible for the death of 89% of those who died testing positive for coronavirus ("morti per"), with the remaining 11% of deaths being caused by some other disease or pre-existing condition ("morti con"). Even excluding comorbidities such as diabetes and hypertension, which are very common, 18% to 28% of the deaths they analyzed were still directly caused by COVID. I'm not sure this can be brought down to 12% without being exceedingly selective (e.g., people aged 0-30 without hypertension, diabetes, respiratory problems... - I know I'm 27 and I wouldn't qualify). > - Sono state analizzate le informazioni riportate dai medici in 4.942 schede di morte, di soggetti diagnosticati microbiologicamente con test positivo al SARS-CoV-2 (il 15,6% del totale dei decessi notificati al Sistema di Sorveglianza Integrata ISS fino al 25 maggio). Nelle schede di morte sono certificate, oltre a COVID-19, quelle condizioni e malattie che hanno avuto un ruolo nel determinare il decesso. > - COVID-19 è la causa direttamente responsabile della morte nell’89% dei decessi di persone positive al test SARS-CoV-2, mentre per il restante 11% le cause di decesso sono le malattie cardiovascolari (4,6%), i tumori (2,4%), le malattie del sistema respiratorio (1%), il diabete (0,6%), le demenze e le malattie dell’apparato digerente (rispettivamente 0,6% e 0,5%). > - La quota di deceduti in cui COVID-19 è la causa direttamente responsabile della morte varia in base all’età, raggiungendo il valore massimo del 92% nella classe 60-69 anni e il minimo (82%) nelle persone di età inferiore ai 50 anni. > - COVID-19 è una malattia che può rivelarsi fatale anche in assenza di concause. Non ci sono infatti concause di morte preesistenti a COVID-19 nel 28,2% dei decessi analizzati, percentuale simile nei due sessi e nelle diverse classi di età. Solo nella classe di età 0-49 anni la percentuale di decessi senza concause è più bassa, pari al 18%. Televideo source: http://www.televideo.rai.it/televideo/pub/view.jsp?id=888&p=101 http://www.televideo.rai.it/televideo/pub/view.jsp?id=888&p=... ANSA source: https://www.ansa.it/canale_saluteebenessere/notizie/sanita/2020/07/16/studio-istat-iss-covid-causa-diretta-di-morte-per-l89-di-positivi_dda57022-331d-4142-a163-6966e3e87916.html https://www.ansa.it/canale_saluteebenessere/notizie/sanita/2... The ISS-ISTAT report itself: https://www.istat.it/it/files/2020/07/Report_ISS_Istat_Cause-di-morte-Covid.pdf https://www.istat.it/it/files/2020/07/Report_ISS_Istat_Cause...