Putting this into perspective: in roughly a single month, we've already doubled the 7-year average annual number of flu deaths.

Project those C19 stats out across the whole population, and assume (generously) herd immunity effects at 70% that halt the progression, and you're looking at mortality rate that swamps that of all wars, terrorist attacks, and natural disasters faced by the US for the last 50 years, right?

Put differently, unchecked and with that extrapolated fatality rate, we'd be looking at something like 53 years worth of flu deaths at the average annual flu death rate? (Again, I'm just averaging the last 7 years for which we have flu stats; they range between ~12k-45k.)

(Happy to have my math corrected).

I am really surprised how often I have seen "it is just the flu" type rhetoric upvoted on HN. I thought this was an intelligent, technological, scientific, and logical community. I don't see how anyone can square the idea that it is just the flu with evidence like what you posted above. COVID-19 has already accumulated a higher death toll in the US than any flu in over 40 years (I'm sure it is longer, but that is as far back as I can find data on with a quick Google search). That is all with basically the entire country under stay at home orders. There would be hundreds of thousands if not millions of deaths if we treated this like the flu.

Let me provoke some more: I predict that, by the end of this year, the number of total deaths in people over 85 will be about the same as last year.

I'm using the fact that most deaths from this virus occur in people who are already ill from other conditions. Frail old people are much more vulnerable to succumbing to infections.

Last year, about 13.5% of people over age 85 in the USA died [1]. That high death rate means that, at some point in the year, 13.5% of those people fell very ill, and had only months to live. This year, without the pandemic, a similar proportion of people would be close to death. If they caught the virus that causes covid-19, their death would be hastened. However, they would have died anyway, this year.

I know it's callous to say "they would have died anyway," since death is a terrible thing. However, humans have a limited lifespan, for some mysterious reason, and at some point our bodies start to betray us: we develop heart disease or cancer, and our immune systems become less effective. It's as if our bodies are lowering their defenses, almost inviting death. Often, along comes a flu, to finish the job. This year's angel of death is a different respiratory virus, not influenza.

1. https://www.cdc.gov/nchs/products/databriefs/db355.htm

[1] - https://www.cdc.gov/nchs/nvss/vsrr/covid19/index.htm

CDC data puts the US flu fatality rate at 0.1% (up to 20-60,000 deaths for 45,000,000 annual infections ~= 0.1%) and serology studies are putting COVID at around 0.3% -- not adjusting for comorbidities. So it's worse than the flu overall, but massively skewed older so for young and especially young and healthy people it's not materially different than the flu. For kids, its better than the flu -- which kills both the young and the old.

And yeah, the flu kills seemingly young and healthy people, too, we just don't really talk about it.

For old folks, it's much worse. We need to keep them safe, but that doesn't mean we need to freak out about children catching it when they're gonna be just fine. [1] Remember 99.2% of the folks who died in Italy averaged 80.5 years old with an average of 3 comorbidities. Zero health professionals under the age of 49 died in Italy despite constant and repeated exposure. [2]

Not everyone is equally affected by COVID and for the overwhelming majority (99.7% of all people), it's by no means a death sentence.

I strongly suggest you read [2] for perspective. I know all the breathless screaming on television has colored your perspective on all this, and yeah, it's not great. However, allow data to be your guide. We're gonna be just fine.

[1] https://www.wsj.com/articles/consensus-is-emerging-that-chil...

[2] https://www.cebm.net/covid-19/global-covid-19-case-fatality-...

Also the 0.3% number you are referencing is being questioned as other serology studies have shown it to be 2x-3x higher. The reality of the situation is that we don't know the exact fatality rate of COVID-19. What we do know is that it is at least a few times more deadly than the flu and potentially an order of magnitude more deadly. We also know that the coronavirus is much more infectious than the flu. Like previously mentioned, we have already passed the worst flu numbers even while most of us are under a stay at home order. It seems like no matter how you look at the data, the only way in which COVID-19 is not worse than the flu is how it impact the very young. COVID is more deadly in basically every other way. I honestly don't see what people gain from continuing to downplay this disease in relation to the flu.

That's fair, although age matters way, way more, so we can elide the bit for future re-readings.

> Also the 0.3% number you are referencing is being questioned as other serology studies have shown it to be 2x-3x higher. The reality of the situation is that we don't know the exact fatality rate of COVID-19. What we do know is that it is at least a few times more deadly than the flu and potentially an order of magnitude more deadly.

I'm not just referencing the Gangelt data, but also the Vo data, and the Santa Clara data (which is very much in question now). That was also approximately the IFR for the Diamond Princess (0.85%) but of course Diamond Princess skewed much older so likely that IFR represents an upper bound. There are precious few population studies for us to reference, but they're far more valuable for making public health decisions than the ratio of people who go into a hospital as compared to come out. That only defines the numerator, without defining the denominator.

I'm also basing it on the estimates during the pandemic that H1N1 had a CFR of 0.1-5.1% from country to country, and landed on an IFR of 0.02%, or one-fifth the lowest estimate. [1]

> What we do know is that it is at least a few times more deadly than the flu and potentially an order of magnitude more deadly.

Not for all demographics. It's demonstrably less fatal for children, which the flu does not spare, and approximately the same for folks under 49.

> Like previously mentioned, we have already passed the worst flu numbers even while most of us are under a stay at home order. It seems like no matter how you look at the data, the only way in which COVID-19 is not worse than the flu is how it impact the very young. COVID is more deadly in basically every other way.

There's reasons you shouldn't compare COVID's current numbers to an average flu season. One being that COVID's new so none of us have immunity. It's also more virulent. It's affecting a similar size population but just all at once instead of spread out, so it's impact is concentrated in time. That makes it not particularly meaningful to project out. Not to mention that flu number is after a vaccine and after a lifetime of exposures, so likely doesn't represent an apples to apples comparison. If the flu vaccine is 10-60% effective, you should probably multiply the number of expected flu cases by up to 2 - or more. Think of it more like the first time someone caught the flu.

Further, the flu changes regularly, and comes back at us repeatedly, and COVID may well not -- early indications are promising in that regard.

I'm not trying to downplay anything, I'm trying to synthesize the data we do know and not fall into the panic that is dominating the discourse. We shouldn't make big decisions while scared senseless, we should allow the data to dictate as much as possible.

Saying this disease is a "death sentence" for an asthmatic 20 year old is panic. It's just not that. 17% of hospitalizations involved asthmatics in the month of March [2] while 8% of the US population has asthma. So I guess your risk is doubled. From 0.3% to 0.6% without adjusting for age. That is not a death sentence.

[1] https://www.cebm.net/covid-19/global-covid-19-case-fatality-...

[2] https://www.cdc.gov/mmwr/volumes/69/wr/mm6915e3.htm#T1_down

You are not being consistent in your arguments. You are criticizing me for not making an apples to apples comparison and in the same paragraph you are doing exactly that by talking about all the preventive measures that reduce flu deaths while ignoring that the country is currently shutdown to prevent COVID deaths.

You are cherry picking statistics to support your point while ignoring data that conflicts with it. For example the data coming out of New York suggests a higher fatality rate and the numbers you are using from the Diamond Princess are either out of date or flat wrong.

You are both putting words in my mouth and pretending I didn't say things that I did. No one here ever said "this disease is a "death sentence" for an asthmatic 20 year old". I already admitted that the only group that has less to fear from COVID than the flu is the very young, yet you are still bringing up that point when I am talking about overall fatality rates.

Apologies, what I was trying to say is that comparing COVID now to an average flu isn't an apples to apples comparison even if the numbers land where they are now, for all those reasons, some of which are non linear. This is including the ones you mention. The flu has a vaccine, and plenty of past exposures. COVID has the shut-down -- though it also has huge virulence and asymptomatic transmissions that appear to be ongoing.

> You are cherry picking statistics to support your point while ignoring data that conflicts with it. For example the data coming out of New York suggests a higher fatality rate and the numbers you are using from the Diamond Princess are either out of date or flat wrong.

Indeed the Diamond Princess number was out of date. I'm suggesting that there's a broad spectrum of CFR estimates (0.08% to 15%) and there's reasons neither is representative.

> You are both putting words in my mouth and pretending I didn't say things that I did. No one here ever said "this disease is a "death sentence" for an asthmatic 20 year old".

GP, who I realize now is not you said: "If asthma or hypertension have suddenly become a death sentence, that is a very big deal;..."

It'll be interesting to see where the dust settles.

Time will tell. Not saying one way or the other just that the worst case you've illustrated is far from a foregone conclusion.

This is a unfair way to respond to [alejo's] comment. The fact that you have responded in this way to this thoughtful response suggests to me that the problem may lie in the way you are reading, understanding, and internally labeling other people's views.

Honestly, I think we don't do nearly enough comparisons between disease mortality and other causes of mortality. A "War On Cancer" would have done us far more good than the wars we declared in the 2000s would have.

I'm not saying that wars on "terror" are a good thing. It's just that cancer is a very tough problem.

the more accurate exapolation would be relative to the first 4 months of every other studied emerging disease, considering the estimated errors of each of the important variables. you'd fit the various current studies on corona/covid, considering the high uncertainty as (large) estimated error, and providing a range of future outcomes with probabilities on them.

instead, you're asserting your worst possible scenario as the most likely outcome.

~~~~~~~~~~~~

now on the idea that we're entirely too blasé about preventable mortality in general, i agree. if we rigorously and intelligently attacked auto accidents, if we addressed cardiopulmonary diseases caused by pollution and poor habits, and if we regularly shut down the economy for the past decades because we want to reduce the hundreds of thousands of contagious disease fatalities every year and had the socioeconomic safety nets in place for that all along, then the current actions would seem more reasonable.

instead, the response is out of proportion with our own past actions because of panic and fear. it happened with 9/11, mass shootings, fukushima, and everything else. the herd is not good at reasoning about rare and novel mortality events.

That's not generous, or not generous, it's just a ballpark guess based on the best (conflicting) data we have so far. You're also discounting the fact that with every individual to gain immunity, the transmission rate is reduced. You're also extrapolating from what may be the worst possible phase of this virus. As time passes we develop better treatments and medications, we have more necessary equipment where it needs to be, we have better early detection and so on.

"...you're looking at mortality rate that swamps that of all wars, terrorist attacks, and natural disaster"

Lots of things swamp those numbers. For example, the flu.

I haven't looked it up, but 55,000 flu/pneumonia deaths a year is about 153 a day, compared to COVID-19 running near 2,000 a day:

https://public.flourish.studio/visualisation/1727839/

This visualization is from before the acknowledgement that COVID-19 deaths were only counting confirmed positive / presumed positives from the earlier set of CDC symptoms, not the 3x larger set of symptoms revised last week.

Also doesn't account for second order deaths such as those with other health problems not seeking treatment from an overloaded system, or providers in the system committing suicide from PTSD.

Taking into account the lack of all immunity to and lack of vaccine for COVID to date, and just how much more virulent COVID is (therefore front loading the disease burden) in the full course of time the flu is almost certainly going to take a much, much larger toll.

It's killed 600,000 people per year, each and every year you've been alive worldwide. So far, COVID is at 1/3 of a year's worth of flu deaths -- let alone a lifetimes -- and are approaching 25%+ of the population of some high profile cities infected. Once it gets to 70% or so, it's most likely going to burn itself out due to herd immunity.

"It's killed 20-60K people per year, each and every year you've been alive in the US. So far, COVID is at about a year's worth of a bad flu deaths in the US -- let alone a lifetimes -- and are approaching 25%+ of the population of some high profile cities infected. Once it gets to 70% or so, it's most likely going to burn itself out due to herd immunity."

You calculate herd immunity % from R0, and it's odd to decide 70% is not generous and ballpark guess because you think the data "conflicts" - there's no "conflicting" between values in a range, go ahead and cherry-pick the lowest R0 you've seen and find out what the herd immunity % required is based on that (higher than 70%)

Won't even touch you trying to pretend this is the same as the flu, plenty of other threads address that.

I've made it clear already, but just to reiterate, this is one of the worst comments I've seen on HN in a while.

Er, yes.

Project those C19 stats out across the whole population, and assume (generously) herd immunity effects at 70% that halt the progression, and you're looking at (a) mortality rate that swamps that of all wars, terrorist attacks, and natural disasters faced by the US for the last 50 years, right?

Yes. "Flattening the curve" just means keeping the hospitalization rate below what the medical system can handle. The curve stays flat near the peak rate. See the curves at [1]. The Financial Times has been updating those daily since early March. Some countries show a decline, but other than China and South Korea, which did heavy testing and lockdown, most countries peak and decline a bit, then level out above 50% of peak.

That continues until "herd immunity", either via vaccine or exhaustion of uninfected people. That's somewhere around 70-80% for this.

The vaccine situation is looking up. At least two vaccines are already beginning clinical testing for effectiveness. One from Oxford, one from Germany.

[1] https://www.ft.com/coronavirus-latest

https://blogs.sciencemag.org/pipeline/archives/2020/04/23/a-...

It includes additional promising candidates.

As I understand it, and I'd be happy for expert correction: lockdown has additional benefits beyond curve flattening; it also accounts for decreased transmission, which can get us to a point where we can track and suppress outbreaks individually while opening up the economy. That's essentially what seems to have happened in South Korea.

Influenza has a lower R0 than SARS2, and it absolutely kills people. Worth destroying our economy over, all by itself? Ehhhh... I mean, we could have done this at any time, and we didn't, so no.

Nice side effect that translates directly to lives saved? Absolutely.

Does it offset other probable side effects, such as increased heart attack mortality from patients being unwilling to go to the hospital, or increased suicide risk from sudden business failures and unemployment?

No idea, like, none. Biostatisticians are going to be crunching 2020 for the rest of the decade.

So then: if you're going to take on a massive economic hit just from altered consumer behavior, does it make sense to get the worst of both possible worlds by also relaxing regulatory constraints that are demonstrably saving lives?

It's not my stance. This is more narrowly-focused than that: we could end the flu season early, every year, by imposing lockdown from January 1st to February 15th.

But we don't, and this isn't the flu, it's a novel disease which kills many more people by even the most conservative estimate.

Still, as a side effect of the eminently rational (and economically painful) precautions which we've taken against it, ending the flu season a couple months early is a nice bonus.

All the weirdness and horrors of "normal" diseases are now in the spotlight due to media attention on covid-19.

I have a feeling that in 2120 people are going to look back at us as idiots for missing a lot of stuff that will seem obvious in retrospect.

I had lots of complications the last year with a common flu. My joints and some internal organs became inflamed because of the immune response and I could not move and was very painful.

This year I got covid early from meeting a person coming from Shanghai. I thought it was a weak flu.

Now I give positive in antibodies test.

"Every other virus" doesn't behave the same in this regard. The viruses that cause colds, many of which are also coronaviruses, are pretty adept at hitting people throughout their life.

Hopefully it's not an issue here, but it's completely unsurprising that people would want to actually research it instead of just declaring all viruses are the same.

The only fools here are the people whose impatience leads them to declare absolutes ahead of times instead of taking preventive measures while the research plays out.

https://www.npr.org/sections/goatsandsoda/2020/03/20/8190384...

> Researchers do know that reinfection is an issue with the four seasonal coronaviruses that cause about 10 to 30% of common colds.

> This study provides evidence that re-infections with the same endemic coronavirus are not atypical in a time window shorter than 1 year and that the genetic basis of innate immune response maybe a greater determinant of infection severity than immune memory acquired after a previous infection.

http://www.columbia.edu/%7Ejls106/galanti_shaman_ms_supp.pdf

Even if you recover, there's not a lot of evidence now as to what the duration of immunity will be. With this spreading through the world at different paces, it means that, as commonly predicted, we will see continuous re-emergence as we do with other cold viruses and influenza.

Source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2851497/

This hypothesis ("SARS-Cov19 does not confer immunity") falls roughly under the same category as intelligent design: the claim is so exceptional that it's a priori unlikely to be found true, and laypeople are taking normal scientist hedge-speak and exaggerating the meaning in their heads (e.g. "evolution is just a theory").

There was never really any question to anyone with any serious scientific training that humans would develop some degree of immunity to this virus. That's why people were saying it was more likely to be assay error from the start. The only open question here is how long that immunity might last. Even then, it's fairly clear that short-term (~months to a year) immunity is highly likely.

(Note: I have a doctorate in biochemistry with a specialty in molecular biology, and I've studied virology and immunology. I'm not just tossing off uninformed opinions.)

No. As the person you're responding to pointed out: "The viruses that cause colds, many of which are also coronaviruses, are pretty adept at hitting people throughout their life."

You can get temporary immunity that lasts for a year or so and still be able to get reinfected multiple times in your life.

Yes. I'm talking about the question of immunity, not whether or not it's possible to get re-infected by new strains of the same virus, or whether or not immunity fades over time. It would be unlikely that this virus didn't confer at least short-term immunity to the host.

The viruses that cause colds primarily re-infect because they mutate at a very high rate. We aren't seeing that with this particular virus.

https://translational-medicine.biomedcentral.com/articles/10...

This is not claimed anywhere in the paper you linked to. They show a non-synonymous mutation in RNA-dependent RNA polymerase, and hypothesized that it could impact drug resistance.

> We found that viral strains with RdRp mutation have a median of 3 point mutations [range: 2–5], whereas viral strains with no RdRp mutation have a median of 1 mutation [range: 0–3] (p value < 0.001, Mann–Whitney test). The different distribution between the two groups relative to the number of mutations is statistically significant (Fig. 4).

So call the null hypothesis "there will be some level of immunity." But then you see some people testing positive again, and it's reasonable to want to know why. I didn't see people officials leaping to planning for zombie apocalypse "we're all gonna get it over and over forever" scenarios, I saw people saying we needed to dig deeper.

A lot of people in this discussion are refuting a straw man idea that everyone had decided "clearly you get no immunity at all".

I have been following the statements of experts ever since this myth made headlines.

And despite what the headlines said, the opinions of experts has always been clear. That being that these "reinfections" were the result of dead virus material being detected.

So, no it is not reasonable to have ever believed this, if you had listened to the experts.

Yes, but you're almost always immune for at least a year. By then, we should be getting close to a vaccine.

And when you do get one of those viruses again, it's significantly less severe the second time.

https://www.virology.ws/2009/12/28/reinfection-with-2009-inf...

How does this follow?

Here is the thought process, though. Typically when you get an infection, your body builds antibodies that can attack the pathogen. The next time the pathogen shows up, your body kills it before you get sick. This immunity doesn't always last for a lifetime, but there is pretty much always some immunity. With COVID-19, early reports showed that this wasn't happening; people got sick, recovered, and then got COVID-19 again. What the news article says is that the test was wrong; when they tested positive the second time, the test was just returning incorrect results (because of limitations of the PCR technique; it's testing for certain RNA, not for the existence of an infectious live virus).

So to say that this means you can be immune to SARS-CoV-2 is not strictly correct. What it says is that there is no reason to think that this virus is any different from many of the other viruses that we already know about. (We do know about viruses that you can't be immune to, like HIV. But those viruses explicitly attack the immune system machinery that is supposed to kill them. We have no evidence that SARS-CoV-2 does that.)

Are they all false? Don't know. There's hope, but not yet evidence, so far as I know.

You mean there's evidence but not proof.

Can you provide some sources?

My understanding was there were many different tests being used ... so on what basis are you making that generalised statement.

> When you recover from it, you don't still have it, even if you test positive. //

A plausible explanation for specific tests doesn't make this a universal truth; so presumably you are relying on more information? Also, there being an explanation for positive tests based on dead virus, doesn't mean all post-infection positive tests are due to dead virus; that's a massive assumption.

.. assuming the virus can not hide (and be able to leave) where they can't penetrate - like in the eye ?

.. and with weak immune system the outcome may be (how?)long undecided as the antibodies are not being able to fight the virus fast enough to finish it definitely?

Do we know for sure that Covid19 in N.Europe isn't sufficiently mutated over that in Australia, say, that secondary infection is possible (even if recovered people's antibodies recognise it somewhat)?

What I've heard from experts in the field, but I don't have sources, is that the type of RNA virus is different from influenza and doesn't mutate as quickly - which is potentially good news for those hoping for a vaccine.

Well, it may appear to be the case so far but afaik that hasn't been rigorously confirmed, and this article doesn't imply immunity to live virus upon reexposure.

>but that doesn't mean the virus is some sort of strange monster

Except there are dozens of papers being publishes about a range of effects - damage to a multitude of organs, virus fragments found in brains during autopsy, high incidence of clotting/stroke. By many SME accounts it is very much a strange monster. Also recall the documented rapid progression from asymptomatic to dead.

>It's extremely infectious, and somewhat more lethal than influenza

So one of these numbers (CFR) is about the same, but the R0 is anywhere from 2-4 times larger, and the critical rate appears to much higher than the typical flu. So the response of "terror" isn't totally unwarranted.

>When I first heard about those patients who had recovered from the virus, and yet were still testing positive, it felt like a mistake. I'm glad it's been cleared up.

Agreed. And another reinfection case occured in an immunocompromised patient. So it's looking like the infection doesn't become chronic or dormant.

That’s not how HSV works. That’s not how hepatitis works.

There you go

Moreover, recent literature continues to confirm that you have protective immunity after recovering from covid like any other coronavirus. This is the basis for convalescent plasma therapy

It's interesting to read "it's extremely infectious" and "more lethal than influenza" followed by "that's all", not to mention the missing bit about not having a vaccine!

If we saw a massive outbreak of influenza things would be extremely different. We have treatments for influenza (tamiflu), we have vaccines, we have understanding, and a large portion of the population that is already immune.

The similarities are... they are both viruses. Not sure why anyone would compare the two except to try to explain how different they are.

Probably not.

It seems [1] that in humans and animals, there's no long-term immunity arising from infection.

[1] https://www.quantamagazine.org/what-can-other-coronaviruses-...

I always suspected those were mostly due to false negatives from PCR tests. Probably only a tiny minority of people had not enough immune response for them to become reinfected.

And various other edge case permutations like that.

Admittedly, I don't really keep up on the status of HIV, maybe something has changed.

I am not sure if anybody gets infected, then clears the virus on their own.

My understanding is that most evidence points to order of magnitude more lethal even when the hospitals aren't overflown.

Regarding "testing negative then positive" the experts already expected that there were no real reinfections.

If those mortality estimates were based on the number of positive test cases, then they're very likely overblown. Recent seralogical studies indicate that the number of positive test results are likely a small fraction of the total number of cases, and likely the most severe subset at that. None of this is to say that it's less lethal than influenza, only that the situation is developing rapidly and any given mortality estimate may well be outdated.

Covid19 figures in the UK are 26,000 [hospital] deaths for 165,000 confirmed infections. Even if we assume the entire population is infected ... that makes it a minimum of 10x as deadly as flu. We're at a slowly falling ~5000 deaths per day; when care-home and other Covid deaths get added ...

I think perhaps you're down-playing it a little too much.

No: the antibody tests were done with the expectation that the total number of the infected would be much higher. The tests have such false positive rates that with the results actually observed the tests in most of the cases say nothing -- in most of the locations the numbers would be similar even if nobody would be a real positive. And there where the results are above these "noise" thresholds we also have a big numbers of deaths which do confirm my claim of order of magnitude deadlier than flu.

Even observing the pure death statistics we can see that, even in spite of all the measures actively taken throughout the world. Without them it's obvious it would have been again the difference of some orders of magnitude:

https://www.euromomo.eu/graphs-and-maps/

See also here:

https://www.ft.com/coronavirus-latest

the "Death rates have climbed far above historical averages" graph.

Saying that it would've been a magnitude higher without any of the measures is pure speculation.

https://www.cdc.gov/flu/about/burden/index.html

They don't mention IFR anywhere as far as I can see

But it seems plausible if the CFR is <0.1% as there are both asymptomatic cases of the flu and people don't always go to the doctor even if they have symptoms. It seems like caring about IFR was rare before this pandemic and the lockdown

Early studies treated the Case Fatality Rate as the Infection Fatality Rate, by virtue of the fact that the only people getting tested were people with symptoms. Random testing in Germany has indicated that ~20% of the population was infected or has been infected, and the corresponding death rate yielded an IFR of 0.37%.

Measuring IFR is inherently speculative, unless 100% of the population is administered a test with no risk of false negative or positives. In order to compare the IFR of COVID-19 to the seasonal flu, we'd need to use the same estimation techniques. I've only found a few experts that have done comparisons of IFR rates of the flu and of COVID-19. The most reputable source on this reported an estimated IFR for the flu of 0.04%, but also an estimated IFR for COVID-19 of 0.2-0.3% [2]. This is "many times deadlier" if by "many" you mean 5 to 8 times deadlier. This is nowhere near the multiple orders of magnitude deadlier that was reported initially (IFR rates of 2-3% and above).

1. https://www.cdc.gov/flu/about/burden/index.html

2. https://www.bloomberg.com/opinion/articles/2020-04-24/is-cor...

Actually, the "risk" of false positives or false negatives is dependent on prevalence, 100% is not always required. If you have a test where 5 of 100 cases are false (i.e. 95% times is not wrong), and nobody of these tested is actually infected, you could incorrectly believe that 5% of population is infected even when nobody is, meaning such a conclusion would be completely false.

But if the population is actually already e.g. 50% infected, the same test can "lie" only 5%, giving you 47% or 53% but still being "mostly true" (from the engineering point of view).

So it is important to ignore the test reports as long as they are close to their false positive rate, which they were in a lot of antibody tests done up to now.

Also "false positives" and "false negatives" can lead to wrong handling of the cases, but that's another topic.

Initially you were implying that the IFR of the flu was 0.1% and comparing it to an IFR of 0.4%, this is what I disagreed with strongly

The bloomberg link you're giving is showing that the vast amount of randomly sampled serological studies are showing IFR of 0.5%-1%. And importantly, looking at how the random sampling has been done for each case and ranking by quality, >0.5% results rank at the top (he studies with the biggest sampling problems show the lowest IFR)

0.2-0.3% is by far in the bottom range of estimates

I think the data shows a pretty clear single order of magnitude difference between the IFR of the flu and novel corona. But I agree that >1% IFR is extremely unlikely

Plenty of the >0.5% results have serious sampling problems of themselves. The Dutch study sampled people who were donating blood. The subset of the population that donates blood is could easily have different behavior than the general population: like being more health-conscious and thus less likely to be infected, which would inflate the infected fatality rate.

And when we compare against the flu, we also have to consider that there are different methods of estimating the infected fatality rate. More conservative estimates on infections of the seasonal flu yield higher IFR values. The epidemiologist being cited here is providing relative estimates of 0.04% and 0.2-0.3%. We could go with more conservative estimates on the infected rate for COVID-19, but we'd have to be similarly conservative when measuring the IFR of the flu.

I'm not sure the blood sampling would skew the direction you think it does, healthy individual could just as well mean that they're more likely to have had it since they're less likely to sit at home and stick to a quarantine. But you are right, it definitely counts as a sampling issue and I'll concede that ranking by quality is a lot less clear cut, I should not have been so confident

What you write is disconnected from the article, immunity was discussed more deeply elsewhere on HN already.

At least for some time.

> In other words, covid-19 is just like every other virus, in this respect.

Other coronaviruses don't mutate, yet reinfect lots of people and cause some common cold roughly every two years.

I think we can expect roughly the same from covid19. Peak every two years until we have a vaccine. Then annual covid19 vaccine shot to refresh the immunity.