Infection fatality rate of Covid-19 inferred from seroprevalence data [pdf] • Hacker News Board

Re: Infection fatality rate of Covid-19 inferred from seroprevalence data [pdf]

#271

Post by ceejayoz » Sat, Oct 17, 2020, 11:46 PM UTC

latchkey wrote: post #92

rndgermandude wrote: post #39

Earlier quoted context omitted.

This is a very important point. We don't know enough about long term effects yet. In the same vein, we do not even know if herd immunity really is a thing here, and what thing it would be. E.g. the guy from Hong Kong who got reinfected (first time severe symptoms, second time no symptoms) was probably infectious the second time again. Herd immunity only works if the people who are immune are not infections and theref…

Herd immunity has never happened without a vaccine.

https://en.wikipedia.org/wiki/Herd_immunity

> Herd immunity was first recognized as a naturally occurring phenomenon in the 1930s when A. W. Hedrich published research on the epidemiology of measles in Baltimore, and took notice that after many children had become immune to measles, the number of new infections temporarily decreased, including among susceptible children.

(To be clear, though, I don't think the herd immunity strategy is a good one for COVID-19.)

Re: Infection fatality rate of Covid-19 inferred from seroprevalence data [pdf]

#272

Post by JungleGymSam » Sat, Oct 17, 2020, 11:47 PM UTC

just-juan-post wrote: post #241

NelsonMinar wrote: post #155

The author, John Ioannidis, is a discredited researcher on Covid-19. He and his Stanford colleages are responsible for a terribly written antibody study done back in March on residents of Santa Clara County. The statistics in that paper were egregiously bad and seem cooked to meet foregone political conclusions. Some reporting: https://www.buzzfeednews.com/article/stephaniemlee/ioannidis... https://www.mercurynews.co…

Let's take John out of the discussion. On Oct 05 the WHO officially announced there were 750m cases worldwide. On that day their official fatality count was ~1m. This comes out to an IFR of 0.13% Do you have an issues with this statistic as well?

No rebuttals. Just downvotes. That'll teach you a lesson, right??

edit: Oh no! Someone downvoted this comment? Whatever will I do?!

Re: Infection fatality rate of Covid-19 inferred from seroprevalence data [pdf]

#273

Post by jeffbee » Sat, Oct 17, 2020, 11:50 PM UTC

morsch wrote: post #16

Here's another recent meta-study (preprint) on the same topic: Assessing the Age Specificity of Infection Fatality Rates for COVID-19: Systematic Review, Meta-Analysis, and Public Policy Implications https://www.medrxiv.org/content/10.1101/2020.07.23.20160895v... Abstract: This paper assesses the age specificity of the infection fatality rate (IFR) for COVID-19 using results from 29 seroprevalence studies as well as…

Bt it's great news for people who want some turnover in the US Senate.

Re: Infection fatality rate of Covid-19 inferred from seroprevalence data [pdf]

#274

Post by JungleGymSam » Sat, Oct 17, 2020, 11:50 PM UTC

qwerty1234599 wrote: post #93

rossdavidh wrote: post #2

For comparison, from what I've read the typical annual influenza fatality rate is around 0.1% (don't know what the under-70 rate would be). So, this puts covid-19 solidly in the uncanny valley of viral mortality rates; not low enough to be "just like a flu", not high enough to justify shutting down the world. No wonder opinions on it are so divided.

Daily reminder: new respitory diseases come and go, nobody usually cares. What makes Covid different and what caused the shutdowns was when Lombardy alone had 450 deaths a day. No regular influenza could do that. It's not as if we decided out of thin air the virus was dangerous.

That's not happening anymore. Time to be adults, instead of terrified children, and find alternatives to lockdowns and mask-shaming with the occasional assault, and arrests from local governments.

Re: Infection fatality rate of Covid-19 inferred from seroprevalence data [pdf]

#275

Post by ImaCake » Sat, Oct 17, 2020, 11:59 PM UTC

NelsonMinar wrote: post #155

The author, John Ioannidis, is a discredited researcher on Covid-19. He and his Stanford colleages are responsible for a terribly written antibody study done back in March on residents of Santa Clara County. The statistics in that paper were egregiously bad and seem cooked to meet foregone political conclusions. Some reporting: https://www.buzzfeednews.com/article/stephaniemlee/ioannidis... https://www.mercurynews.co…

Ah that explains why this is not a properly conducted meta-analysis. This guy seems to think he is somehow better than all the checks and safeguards put in place to ensure people do meta-analysis properly. I personally found this meta-analysis was done okay, considering it was not properly anyway, and defends itself well. But it is clear this man is pushing an agenda and that the meta-analysis itself is standing on some very shaky foundations.

Re: Infection fatality rate of Covid-19 inferred from seroprevalence data [pdf]

#276

Post by mrb » Sat, Oct 17, 2020, 11:59 PM UTC

butthewho wrote: post #163

NelsonMinar wrote: post #155

The author, John Ioannidis, is a discredited researcher on Covid-19. He and his Stanford colleages are responsible for a terribly written antibody study done back in March on residents of Santa Clara County. The statistics in that paper were egregiously bad and seem cooked to meet foregone political conclusions. Some reporting: https://www.buzzfeednews.com/article/stephaniemlee/ioannidis... https://www.mercurynews.co…

This is a peer-reviewed article published by the World Health Organization, not a press release from an individual Stanford lab. Could you please enumerate your specific concerns with this study?

Ioannidis has been trying to publish this study for a while. He frequently updates the preprint on medrXiv and others reviewed it and criticized it extensively over the last few months.

Here's a review by an actual epidemiologist pointing out specific flaws: https://mobile.twitter.com/GidMK/status/1316511734115385344

In a nutshell he uses clearly inappropriate samples to estimate population prevalence, and inexplicably ignores multiple large high-quality samples that contradict his findings.

Re: Infection fatality rate of Covid-19 inferred from seroprevalence data [pdf]

#277

Post by dimitrios1 » Sun, Oct 18, 2020, 12:02 AM UTC

Mountain_Skies wrote: post #260

mkolodny wrote: post #226

Earlier quoted context omitted.

My main concern with this study is that it uses reported Covid-19 deaths to infer the fatality rate. According to the paper, the fatality rates in the US are far higher than the rates in China and India. While the inferred fatality rate in the US is as high as ~1.3% (Louisiana), in many other places like China outside Wuhan and in India, the inferred death rate is close to 0.0%. I highly doubt that's actually the cas…

Speaking of Pakistan, a friend who is a pediatrician there said that childhood vaccinations have declined by 50% since the start of the pandemic, possibly because parents are too scared to go to a medical facility to have it done out of concern of being infected by covid19. The secondary impacts from the pandemic and our response are going to start becoming more apparent.

Suicides, hunger + starvation due, cancers and other diseases or illnesses not caught early, domestic abuse incidents, civil unrest (people get antsy when they have nothing to do). I'd say they are already pretty apparent, people are just fine to have their heads in the sand.

Re: Infection fatality rate of Covid-19 inferred from seroprevalence data [pdf]

#278

Post by gruez » Sun, Oct 18, 2020, 12:04 AM UTC

dabbledash wrote: post #234

jointpdf wrote: post #228

Earlier quoted context omitted.

I disagree strongly with the notion that mitigating risk of harm to others via awareness and simple action somehow makes society less human. I live in DC, a notoriously cold and closed off city as far as social interaction goes. Mask usage (even outdoors) is close to 100% here. When I see someone wearing a mask, I’m filled with a sense of warmth and camaraderie—that person cares about me and others. Someone else may…

It goes far beyond masks. No hugging. No shaking hands. No singing. No bars. Masks and plexiglass for kids who are lucky enough to actually get school. Not having funerals for the dead. We can put up with these things in the short term to protect the vulnerable, and you’re right that people’s concern for others speaks well of them. But the logic some people use to support these things veers into territory where it so…

>But the logic some people use to support these things veers into territory where it sounds like they would be willing to make (or demand) any sacrifice for literally any increase in safety. I worry about how that will play out long term.

Seems like you're strawmanning/shifting the goalposts, considering your initial comment says "I would just go back to living like I lived before", which would suggest you would take zero protective measures.

Re: Infection fatality rate of Covid-19 inferred from seroprevalence data [pdf]

#279

Post by ImaCake » Sun, Oct 18, 2020, 12:04 AM UTC

morsch wrote: post #16

Here's another recent meta-study (preprint) on the same topic: Assessing the Age Specificity of Infection Fatality Rates for COVID-19: Systematic Review, Meta-Analysis, and Public Policy Implications https://www.medrxiv.org/content/10.1101/2020.07.23.20160895v... Abstract: This paper assesses the age specificity of the infection fatality rate (IFR) for COVID-19 using results from 29 seroprevalence studies as well as…

Your linked study is much better than Ionnadis' one! This one actually followed the proper protocol for a meta-analysis for one. The results here are better stratified, the data more solid (because they did their meta-analysis properly), and the justifications better.

For those not familiar with biostatistics and meta-analysis. There is a standard and well defined pathway for doing a meta-analysis. It exists for a reason. Namely that it is really easy to misinterpret the results of aggregated studies if you do it wrong. The fact that the Ionnadis' article does not follow these procedures is enough to disregard it and instead focus on understand the studies that have done proper meta-analysis.

Re: Infection fatality rate of Covid-19 inferred from seroprevalence data [pdf]

#280

Post by mrb » Sun, Oct 18, 2020, 12:04 AM UTC

dimitrios1 wrote: post #277

Mountain_Skies wrote: post #260

Earlier quoted context omitted.

Speaking of Pakistan, a friend who is a pediatrician there said that childhood vaccinations have declined by 50% since the start of the pandemic, possibly because parents are too scared to go to a medical facility to have it done out of concern of being infected by covid19. The secondary impacts from the pandemic and our response are going to start becoming more apparent.

Suicides, hunger + starvation due, cancers and other diseases or illnesses not caught early, domestic abuse incidents, civil unrest (people get antsy when they have nothing to do). I'd say they are already pretty apparent, people are just fine to have their heads in the sand.

The only data I've seen (Peru) suggests that lockdowns significantly 𝗱𝗲𝗰𝗿𝗲𝗮𝘀𝗲𝗱 homicides, suicides, traffic accidents, and other accidental deaths. I wonder if the same effect exists in other countries: https://mobile.twitter.com/zorinaq/status/131484402784650035...