https://www.buzzfeednews.com/article/stephaniemlee/ioannidis... https://www.mercurynews.com/2020/05/24/coronavirus-research-...
Could you please enumerate your specific concerns with this study?
The methodological criticisms of the paper that you’re making, while relevant, in no way invalidate the work. While the estimate may have been off by a factor of 2-3, overall, that’s well within the margin of error for a study of IFR on a small sample.
This comment, downthread, cites a paper which finds similar estimates to Ionnadis’ original paper:
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 case. It seems much more likely that Covid-19 deaths have been under-reported in China and India.
Anecdotally, my Pakistani friend called Covid-19 a "rich-person" disease because only rich people in Pakistan can afford a Covid-19 test. That may account for the lower number of reported Covid-19 deaths in poorer places.
Traffic accidents being down makes complete sense -- nobody is going anywhere.
The CDC says there is an increase in suicide ideation https://www.cdc.gov/mmwr/volumes/69/wr/mm6932a1.htm
The PAHO says the pandemic exacerbates suicide risk factors https://www.paho.org/en/news/10-9-2020-covid-19-pandemic-exa...
There are some other studies that I don't necessarily know what to make of that already say they are increasing.
People are afraid to go into a hospital. From March until September (for some states), people were being turned away because their routine screenings and exams were not seen as essential. Screenings and exams that can catch lumps, polyps, etc. Routine medical care is down across the board. Dental health is suffering.
For food, global supply chains have been interrupted and brought to a halt. These have months-long reverberations that have yet to be fully recovered. The people who suffer the most from supply chain interruptions are the most vulnerable, and those that live in food deserts. Again, money hasn't been mentioned yet.
But yes, you do have a mass loss of ability to provide, and businesses going under at record pace. There isn't a social safety net in the world you could construct to suddenly handle millions going out of work because of lockdowns.
> There isn't a social safety net in the world you could construct to suddenly handle millions going out of work because of lockdowns.
I disagree. Seems easily doable, given the wealth and capacity of the country. Just a matter of political will.
Why have other countries managed to take care of their workers? Maybe we can ask Canada for pointers?
In any case we know India and Brazil (2nd and 3rd country with highest death toll) are underreporting deaths: https://www.bmj.com/content/370/bmj.m2859 and https://mobile.twitter.com/muradbanaji/status/12844812155635...
https://mumbaimirror.indiatimes.com/coronavirus/news/sero-su...
If someone dies without having already tested positive for Covid-19, in many places they won't be counted as having died from Covid-19. So it's likely that many Covid-19 deaths haven't been counted - especially in places where most people can't afford the tests that check whether they currently have Covid-19.
it's very plausible that it's < 0.1%. Singapore and large parts of Africa or the ME have seen few deaths. Young populations, low rates of obesity, diabetes and other diseases that probably plague regions like Louisiana will change the outcomes quite drastically.
Singapore in particular is probably a high fidelity case. It has one of the highest testing rates on the globe, likely clean data and they've registered 28 deaths on >50k cases.
the very large majority of their cases were in foreign worker dorms. almost none of these workers are over 40, and are generally in good health.
so the demographics are very different.
yeah sure but that's the point. the demographics in much of the world are more favourable. The median age in Nigeria is 18 years, on the African continent 20. Much of Asia is relatively young as well. It's the Western places who are the outlier.
singapore has a more elderly population than the usa. the covid outbreak was largely in foreign worker forms which have a much lower average age than the rest of singapore. singapore is not nigeria. it’s quite an elderly society.
"Locations are defined at the level of countries, except for the USA where they are defined at the level of states and China is separated into Wuhan and non-Wuhan areas."
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.
Your comments are a textbook ad hominem attack.
If you can, post a definition of the terms you use to get more people familiar with the terms and concepts, even if it just boilerplate copypasta.
It warms the cockles of my heart to think we can actually tighten up discussions and get the heart of matters by avoiding rhetorical sillinesses that have been well-known in some circles for over 2000 years.
Maybe a good use for GPT-4+ will be a bot that will identify these rhetorical devices in forum threads.
And yes, it is an ad hominem attack. The basic point may or may not be correct, but the post immediately loses cred for using such a device. OK, now let’s try post hoc ergo propter hoc...
I'm not an expert in the replication crisis but I have to say that as a conceptual framework, it has the problem of pointing a finger at all research - saying "Most Published Research" etc, when the replication crisis is quite concentrated in experimental psychology, complex biomedical causation research and fields with pretty specific difficulties (humans taking simple psychology tests seem to be easily influenced by outside factors, the biochemistry of mammals seems to incredible variations in it's operations). The germ theory of disease doesn't by itself fall into this paradigm (even if humans are complex, the transmission mechanism of a germ is simple). And there's the problem, it's as if Ioannidis decided, "nothing is true, I'll claim what I want".
He's push covid denialism in every conceivable forum open to him and aligned with overtly right wing ideological institutions like the Hoover Institute and the American Institute For Economic Research.
This stuff has been gone over so much by now it has skidmarks on it but I'll dust off the file I keep this stuff:
A nice article describing the similarity of Covid-denial and Climate-change-denial: https://www.yaleclimateconnections.org/2020/04/coronavirus-d...
An article on the effort to spin the Santa Clara tests befor e they were even release: https://slate.com/technology/2020/04/coronavirus-circulating...
What's on with Ioannidis: https://undark.org/2020/04/24/john-ioannidis-covid-19-death-...
Richard A. Epstein of the Hoover Institution, another Stanford Connected "minimizer" https://www.newyorker.com/news/q-and-a/the-contrarian-corona...
A summary of several related minimizers, with a similar program: https://arcdigital.media/what-the-federalist-doesnt-get-abou...
Aljazeera gives a broader discussion, why real scientist would do this: https://www.aljazeera.com/indepth/opinion/coronavirus-herd-i...
Several articles revealed conflicts of interest behind the study such as the CEO of Jet Blue promising lab funding for someone who didn't want their name on the study if they would sign on, or something roughly along those lines (maybe with another layer of indirection).
> his IFR estimate could be off by a factor of 2 or 3,
His early opinion piece I think opined it might be as low as 0.05%, when we had very good reason to believe that not to be the case:
> That huge range markedly affects how severe the pandemic is and what should be done. A population-wide case fatality rate of 0.05% is lower than seasonal influenza. If that is the true rate, locking down the world with potentially tremendous social and financial consequences may be totally irrational. It’s like an elephant being attacked by a house cat. Frustrated and trying to avoid the cat, the elephant accidentally jumps off a cliff and dies.
> Could the Covid-19 case fatality rate be that low? No, some say, pointing to the high rate in elderly people. However, even some so-called mild or common-cold-type coronaviruses that have been known for decades can have case fatality rates as high as 8% when they infect elderly people in nursing homes. In fact, such “mild” coronaviruses infect tens of millions of people every year, and account for 3% to 11% of those hospitalized in the U.S. with lower respiratory infections each winter.
https://www.statnews.com/2020/03/17/a-fiasco-in-the-making-a...
He was one of the big sources of credibility for people that extrapolated that to "just a normal flu".
> If he says the IFR is 0.27% and it is really 0.6% that is only off by 0.33. Anyone who thought it was 1.5% or 3% would've been off by a much higher amount.
Remember he was against most actions at the worst time for people being unprepared and hospitals potentially getting overwhelmed. IFR would increase with overwhelmed hospitals and without the time we've bought to develop better treatments and protocols.
There was lots of fishy stuff going on with the later seropositivity study:
Sample bias: https://www.buzzfeednews.com/article/stephaniemlee/stanford-...
https://www.buzzfeednews.com/article/stephaniemlee/stanford-...
https://www.buzzfeednews.com/article/stephaniemlee/ioannidis...
https://www.nature.com/news/1-500-scientists-lift-the-lid-on...
A domain where results are easily reproducible is settled science, and not what people are researching anymore. Every field moves on till they get stuck on problem areas that hinder progress. Kind of a Peter principle for research. And what's worse, bad studies act as a hindrance since they make an unsolved question look like settled science, and you can't get support for pursuing a competing hypothesis.
And epidemiology seems really complex to me. Every person is unique in many ways and behaving based on effects stretching back over their entire lives.
The emails suggesting a foregone conclusion make Ioannidis look even worse.
Nonetheless, the work stands on its own and should be evaluated as such. You would be forgiven for going in with the prior that Ioannidis has a poor track record in this area - but your prior alone is not enough to dismiss the work.
Well that's simply not true, by definition.
That's like the whole point of science. You can do something and write a report on it, and if I don't believe you I am free to try and reproduce it and publish my own account.
Track record is an excellent reason to be skeptical. Skepticism alone is not enough to invalidate a study.
That someone has engaged in bad faith and bad methology previously doesn't invalidate their findings. It doesn't prove they're wrong. But it makes people justified in ignoring them.
The world is full of, uh, bullshit, full of unjustified claims on this and that. These have to be ignored because otherwise you waste all your time. Being a credible scientists engaging credible research is a reason to take someone out of the this category and pay attention to them. But once someone has discredited themselves as a scientist, they're back in the bullshit category and no one has an obligation to look at their stuff. Sure, maybe their stuff is true, who knows.
I do think that some mistakes should be recoverable from though. Outright data forgery should be career ending event full stop - but the burden of proof should be high.
I already posted a second link for you, from the SJ Mercury News, which you chose to ignore. Perhaps you didn't see it. Here are some more:
https://www.buzzfeednews.com/article/stephaniemlee/stanford-...
https://www.mercurynews.com/2020/04/20/feud-over-stanford-co...
https://twitter.com/carlzimmer/status/1251176233594949632
These are all popular articles (and one tweetstorm) because that's what I have collected. If you spend just moments looking in scientific community discussions you will also find many, many critiques. But really it's that first Buzzfeed article you have to go back to because it establishes the motive for the badly conducted and analyzed study, a political agenda and a foregone conclusion. It's the worst form of scientific corruption.
I'm not sure what to do about it. I know MIT better than Stanford, and fixing the culture would involve firing a big chunk of the faculty. You can't fire tenured faculty, even if you could get people to acknowledge the problem.
Integrity at 2nd and 3rd tier schools is still a lot higher, but we do need to fix incentive structures, or the rot will take over academia.
Softer disciplines (like social sciences) are way ahead of harder disciplines like CS on the rot curve. In CS, it's a lot harder to bake data.
You've never read a machine learning paper, I take it. Those results are by far the easiest to bake.
Similarly your post is an ad hominem, but you didn’t make any arguments based on merits.
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?
https://apnews.com/article/virus-outbreak-archive-united-nat...
The 750mn cases is an estimate based on seroprevalence studies. The 1mn deaths are based on deaths following a positive PCR test. A more appropriate comparison is 1mn deaths over 35mn cases, leading to a CFR of over 2%.
Odd how no one ever quotes that number.