Hacker News new | past | comments | ask | show | jobs | submit login
[flagged] 2009 flu pandemic in the United States (wikipedia.org)
33 points by halamadrid on March 16, 2020 | hide | past | favorite | 76 comments

The CDC's worst case estimate was 214 Million Infected, 1.7 Million Dead


To be clear, this is the worst-case estimate for the current coronavirus outbreak in the US, not the worst-case estimate for the 2009 H1N1 flu outbreak.

I'm not sure how they got 1.7 million dead as the worst case scenario. That implies a mortality rate of 0.7%.

Mortality rates for Covid19/SARS2 are estimated to be closer to 3.5% by the WHO, a 35x jump. At 214 million infected, the worst case scenario should be at least 7.5 million. But I believe the 3.5% estimate also assumes that hospitals are able to provide intensive care for all critical patients.

The US only has 100k ICU beds, and SARS2 has an ICU rate about 10%. SARS2 patients typically need 2-4 weeks, meaning that the entire US healthcare system can only handle about 2 million SARS2 infections per month. In order to keep a mortality rate of 3.5%, the US would need to spread out those 214 million infections over 9 years.

Conclusion: The CDC's worst case estimate of 1.7 million deaths is completely incongruent with their worst case estimate of 214 million infections.

3.5% just doesn't make sense based on any of the case-fatality-rate data we've seen so far combined with the estimates of what proportion of infections are being detected as cases. The UK government's chief medical adviser expects the real fatality rate to be around 0.6%, which is a lot easier to explain with the data than 3.5%.


"The case-fatality risks, when adjusted for a 13-day lag time from reporting to death, were 3.5% in China; 0.8% in China, excluding Hubei Province; 4.2% in the group of 82 countries, territories, and areas; and 0.6% for the cruise ship (Table). Our result for China, excluding Hubei Province, is similar to a previous estimate of 0.9% (95% CI 0.6% - 1.3%) by using a time-delay adjusted case-fatality risk for the same area"

This is the only research data I could find. Do have other data? I have multiple sources citing the WHO estimating 3.5%, but I was unable to find any first hand material from the WHO that said 3.5%.

The information above - from the CDC, which I consider to be biased and underplaying the situation - has 0.6% as the lowest number, and 4.2% as the highest number.

But honestly none of the data I just went digging through felt comfortable.

Another thing to remember is that the case-fatality-rates we are seeing in most areas are because the hospital infrastructure can keep up with the number of infections. If the US gets over 1 million infections, the hospitals will not be able to keep up and mortality rates will spike significantly.

Wait, you're talking about case fatality rates. I thought the above discussion was about the real fatality rate. Obviously CFR is going to be higher because you're excluding all the infections that didn't become cases - which are also the infections least likely to lead to death.

The morbidity rate is around 5%. In locations where they can keep the ICU beds from being overwhelmed, the mortality rate is low. In locations where the ICU beds are overwhelmed, the mortality rate goes up considerably. It all comes down to how well the system can respond to the load of people needing substantial medical care.

It doesn't "all come down" to that, the testing regime and thus the proportion of infections that are actually detected is an enormous factor.

Without quarantines, COVID-19 will be 280 million infected and 6 million deaths in the US alone.

I don't know why you're being downvoted. Anyone disagreeing with your figure should post a refutation.

I'm shocked there are so many people treating covid-19 as if it's just the flu. Our medical system is going to be strained to the breaking point. It's a big fucking deal.

There's a dog park with booze just outside my window, and it's packed to the brim with hipsters day after day. None of them care. The bars are overflowing with the St. Patrick's Day crowd.


Why are so many people nonchalant? Can't they imagine how this situation plays out?

As someone that is trying to wade through all the misinformation and hysteria, these kinds of comments that throw out numbers with zero sources and zero commentary are useless at best. If you're going to post statistical information (e.g. the virus has a 2% fatality rate) then link to the organization or study which said so.

There's no statistical information in the comment you're replying to...

The comment I'm replying to is a response to the one above it and assumes a certain fatality rate, which is presented with no sources.


Agreed! Thanks to Twitter I caught the severity of this "early" and about two weeks ago had my WTF moment. Felt like I was in the middle of an eartquake for a good hour. A switch flipped. For business and personal life I knew that I had to change up just about everything in order to address and ride through this as best as possible.

Here is something I posted a week ago about the social landscape that still holds true unfortunately:

Twitter: oh shit

Instagram: what virus

Facebook: you’re overreacting

In person: toilet paper!!!!

1) Why are you assuming it will be 280 Million and not stop at around 59 Million like before?

2) Why choose a number like 6% when it's currently at 3.8%?

3) Why are we experiencing so much more panic over this virus?

> 3) Why are we experiencing so much more panic over this virus?

Italy's CFR is over 7%. They've implemented triage in Lombardy where they won't even consider someone over the age of 65, because they're out of beds and ventilators and want to ensure patients with the best chances for survival make it.

Amongst those with symptoms, this virus has an 80% hospitalization rate.

This virus is a test, and we're going to be found severely unprepared.

Give me a moment and I'll go dig up my citations.

Edit: I was wrong about the hospitalization rate. It appears to be 20% [1]

One source cites a CFR of 5% in Italy as of two days ago. [2]

[1] https://www.nbcnews.com/health/health-news/people-have-been-...

[2] https://www.livescience.com/why-italy-coronavirus-deaths-so-...

The CFR is much larger than the IFR in places like Italy that lack testing capacity. The IFR is what's relevant when estimating number of deaths if the virus were to spread to the ~whole population.

80% of symptomatic cases needing hospitalization seems very unlikely. That sounds like a biased testing sample that’s not even testing anyone that isn’t extremely sick. South Korean data seems vastly more comprehensive. Most data I’ve seen puts the upper bound on hospitalization rate at 20% which is still insanely high given how high the R0 is.

Avg age of tests in S. Korea are around 25, avg age in Italy around 65. That makes ALL the difference.

Do you have source for people over 65 being rejected?

I read that there is some triage but this seems a lot more extreme than I imagined.

I thought it was a 20% hospitalisation rate?

I think you're right. I can't find the article I thought I had read.

Or perhaps the 80% figure was patients within a certain age range.

H1N1 had an R0 between 1.4 and 1.6 [1]. Let's assume it was 1.6.

Let's assume that the current coronavirus has an R0 of just 2 (estimates vary, most higher than this). Perhaps it is less, but for the sake of argument lets say it's 2.

1.6^10 = ~110

2^10 = 1024 (so ~9x higher)

That's with just 10 steps of transmission. So assuming the R0 of 2 is correct this will spread much faster.

Now consider that people do not seem to show symptoms until at least 5 days after they've been exposed (by some estimates, even longer, like 9+ days). Assuming they are contagious at least part of that time, they are unknowingly spreading this.

As others have mentioned add into it the fact that people who have been exposed to various flu viruses can have some immunity to new ones.

Ignoring all that, just look what's actually happening on the ground in Italy and China.

[1] https://www.ncbi.nlm.nih.gov/pubmed/19545404/

I've heard 40-70%, it's much more contagious 2x more than the flu or h1n1. So 50% puts it more at 150 millions * .035 = 5.25 million baseline.

Now if we run out of hospital beds and we have to pick/choose who lives/dies including car wrecks, and other emergencies the deaths could double or triple.

Not the commenter you're asking, but re: 2, 6/280 = 2.1% for the assumed mortality rate

One big difference being that this is not flu. There is no reason to assume it will behave like flu. Flu is contagious 2 days before symptoms, comparatively limiting it spread to corona which is contagious 5-14 days.

Corona has higher mortality rate in most age groups, including in young people. People who do need ICU stay there for weeks. Unknown number of people have reduced lung capacity after for long.

1. The R0 of the disease sets the percentage at which herd immunity sets in and the virus stops spreading. The flu has a much lower R0 than the coronavirus.

3. Because it's much worse than the 2009 flu on almost every metric. If we treat them the same, there will be two orders of magnitude more deaths.

The death toll is going to rise without proper care (the reason was given). I'd more modest like 35% of population is likely to get it.

> 2) Why choose a number like 6% when it's currently at 3.8%?

GP appears to be using 2% mortality rate

I just felt that we went through something similar in the past and was hopeful that we should be in a better place to handle this new incident? If not where you think we failed and what could we be doing for the next time?

Look at what Taiwan did. They have fewer than 100 cases, despite close contact to China.

Or you could look at Cambodia. A handful of cases despite very close contact with China (and no travel ban affecting China, whereas Taiwan has one). Comparing cases between countries is completely meaningless, because every country has its own testing regime.

You seem to imply that Taiwan is not testing people. That is not true.

You seem to imply that Cambodia is not testing people. I don't think either of us know for sure one way or another.

Case numbers (and thus CFR as well) aren't comparable between countries, period. It's not as simple as "testing people" or "not testing people", each country has a different regime. Most countries won't test random asymptomatic people without a reason to do so. Heck, many countries won't even test symptomatic people unless there is a known link to a case.

I know for sure that Taiwan had an extremely early and extremely thorough response. They started tracking people who traveled to Wuhan by the end of December, when there were hardly any cases there. I have doubts that Cambodia has the systems in place to act similarly, but I'd be happy to hear otherwise.

Oh, I absolutely have doubts as well. But I don't believe any of this affects my point, which is that since the testing regime is different in each country, comparing the number of cases between countries is deeply problematic.

How is this even remotely similar? H1N1 killed less people in Italy in a year than coronavirus killed in a single day.

How did you arrive at 6 million?

That's 2% of 280 million, which is the conservative number for the COVID-19 deathrate

No, it's not the conservative number. The UK chief medical adviser estimates it at 0.6%, which is much easier to support with the available data (case fatality rate and an estimate of what proportion of infections are being detected as cases) than 2%.

2% is not a conservative number. It's expected to come in at 0.1%-0.5%. That's bad enough.

I did not seen 0.1%-0.5% estimate in any official claim. They were all 2% or above.

Chief medical adviser of the UK govt claims 0.6%.

2% or above is only reached in populations that are tested sparsely. In countries that test more like South Korea and Germany* the death rate is much lower, because more infection cases are known. Italy's death rate is currently so high because only severe cases have been tested, because of limited test availability. The death rate will trend lower when more tests are performed. Germany currently has ~6000 infected and 13 death.

* even in Germany you are (at least till very recently) only tested if you show symptoms AND have been in contact with an infected person or have been to a heavily hit region).

What data supports 0.1% - 0.5%?

Numbers from Switzerland, South Korea, Germany... countries that test more people tend to have lower numbers because not only the severe cases are found.

Please provide a link with data

There is virtually no chance that 85% of the population will get covid-19

It is pretty much in line with worst case scenarios in many European countries.

How's that?

What? That is the worst case that governments are planning for / trying to avoid.

Where did you get that 85% number. Most radical I saw so far was 70%

uk speculates at 80%. I've read 85% (for another country) but I don't have time to dig it up.

https://youtu.be/VNqfElehhKc?t=1765 29:25

Obviously very uncertain numbers no matter who you ask.


I'm mostly hopeful because people are taking COVID much more seriously than the flu.

I couldn't get my parents to get a flu vaccine, but now they are doing self quarantine and taking it.seriously.

As long as people are scared enough, they are relatively safe compared to other diseases.

There are 2 things happening at once that seem to be causing a rift in online discussions:

1. It seems like people are severely underrating the flu and always have been. Remember when we closed all restaurants and bars and suspended all sports events when the flu killed an estimated 1 million people in 1968-1969? No you don’t because society seemed to have shrugged that one off. Ditto for many years when the flu kills upwards of 600k worldwide [1] 2. The reported CFR (Case Fatality Rates) across different countries is all over the place causing a lot of panic and overstated mortality rates. Countries like Italy report almost 7% whereas in Germany and South Korea, the CFR makes Covid-19 look like no big deal (Closer to 0.1%). I suspect that’s because Italy is well under counting the true denominator, while countries like SK have tested thousands of people per day making their data much more comprehensive.

There’s something unique about this year’s outbreak that seems to resonate deeply with more people. Ultimately it’s definitely not “just the flu” but it’s closer to that then a guaranteed death sentence.

We may see more damage from the secondary effects (people losing jobs, runs on medicine and supplies causing people to die) than from the virus itself.

[1] https://en.m.wikipedia.org/wiki/Influenza_A_virus_subtype_H3... [2] https://en.m.wikipedia.org/wiki/Template:2019%E2%80%9320_cor...

What I believe unique about this year's outbreak is that it is the first "social" pandemic; where in everyone is connected to a nonstop stream of both information and misinformation regarding its development.

From How the 2009 H1N1 Pandemic Compares with COVID-19:


For example: Why was H1N1 allowed to spread around the world more or less unchecked, while countries are going to far greater lengths to try to halt Covid-19? Why did the WHO call H1N1 a pandemic but not Covid-19? Isn’t 12,469 deaths a lot worse than the 26 that have been attributed to Covid-19 in the U.S. so far?

That last one is the simplest to answer: Covid-19 is near the beginning of its spread in the U.S., and thus cannot be compared with H1N1’s effect over a full year. If the U.S. death toll from Covid-19 is only 12,469 a year from now, that will likely be counted as a great success. The legitimate worry is that it could be many, many times higher, because Covid-19 is so much deadlier for those who get it than the 2009 H1N1 influenza was.

How much deadlier is still unknown, but of the cases reported to the WHO so far 3.4% have resulted in fatalities. That’s probably misleadingly high because there are so many unreported cases, and in South Korea, which has done the best job of keeping up with the spread of the virus through testing, the fatality rate so far is about 0.7%. But even that is 35 times worse than H1N1 in 2009 and 2010. Multiply 12,469 by 35 and you get 436,415 — which would amount to the biggest U.S. infectious-disease death toll since the 1918 flu. Hospitalization rates are also many times higher for Covid-19, meaning that if it spread as widely as H1N1 it would overwhelm the U.S. health-care system.

That’s one very important reason governments (and stock markets) around the world have reacted so much more strongly to Covid-19 than to the 2009 H1N1 pandemic. Another reason is somewhat more hope-inspiring. It’s that public health experts generally don’t think influenza can be controlled once it starts spreading, other than with a vaccine, whereas several Asian countries seem to have successfully turned back the coronavirus tide, for now at least.

Influenza can’t be controlled because as much as half the transmission of the disease occurs before symptoms appear. With Covid-19 that proportion seems to be lower, meaning that even though it’s more contagious than influenza once symptoms appear, it may be possible to control by testing widely and quickly isolating those who have the disease.

Gosh, the memories. Swine flu caused me to black out for a whole day and yet it seemed to bounce off other people as if it were the common cold.

That’s the biggest unknown about the new coronavirus disease for me: how ill am I actually going to get? Flu seems so unpredictable.

the level of viral load in the initial exposure has a lot to do with how sick you will get. If a sick person coughs in your face that’s much worse than just touching a lightly contaminated cardboard box and getting it that way.

Why? Why is this relevant?

It seems the reason for this is to downplay covid-19. As if that outcome is what the whole world is freaking out over.

If anything you should see those numbers and ask yourself. What if it were exponentially worse?

That is also under the assumption that hospitals will not overrun. Which they will.

Don't plan on having any injuries this year.

To the contrary. That one was that bad and everything points to this being significantly worse. That’s a good reason to post it.

This is what I felt when someone pointed me to the article. I noticed people were still going to birthday parties and other social events when they really should be practicing social distancing.

Even a milder flu like in 2009 could reach 59 million and COVID-19 can be exponentially worse. That is really the only take away. We need to discuss this in the open to bring that truth to light. I agree with some of the folks here who said many could see it another way and take light view of the situation. What exactly can make them see it differently though?

Agree, but that is not the takeaway people will take.

"See, told you, just a bad flu." would be the expected response.

I don't think this downplays it at all. If 59 million were infected with H1N1, what's stopping them from being infected with COVID-19, which has a mortality rate at least 10x higher, if not 30x higher?

Nothing. But people don't tend to make that connection. It is obvious that they haven't so far.

Don't think the intent here is to downplay. Just to remember. You decide what you want to do with information shown to you.

Yes, the public does what it wants to do with the information.

And currently a large part of the public thinks it is a regular flu. So showing them an example of something else that was scary that didn't amount to anything (compared to what we are risking now) is helping exactly how?

I felt there are a lot of people who is looking at the data from China and thinking it will stop at around a 100k total infections in the US. There are so many articles that says life is going back to normal. Even Apple opened all their stores.

No one is disputing the higher death rates but somehow has this feeling that this won't be as bad as other flu incidents?

I felt this was a good reminder that what many fear is going to be true. That 60-70% of population is likely going to be infected. Because that is what happened in the past. It's not going to go away in a few weeks.

COVID-19 is nothing like flu. There are still many unknowns about this virus. We are still not sure of its origins. We have no idea of the long-term effects of it. Some of those who "recovered" already in need to have lung transplants, many have severely damaged lungs. It is difficult to predict how many of those who recovered would live long and healthy lives.

If you are scared shitless and panicking - you should be. Protect yourself and your loved ones. Stay home.

> COVID-19 is nothing like flu.

It is precisely a variant of coronavirus flu. Why do you think otherwise?

It isn't any kind of "flu".

Coronaviruses are in a family that typically causes what we call the common cold. Flu is caused by influenza virus, which is a completely separate family of viruses.

I meant that it is incomparable to the seasonal flu.

I came across this tweet from Stephen Schwartz which he posted after the global health security team was disbanded by John Bolton:


“When the next pandemic occurs (and make no mistake, it will) and the federal government is unable to respond in a coordinated and effective fashion to protect the lives of US citizens and others, this decision by John Bolton and Donald Trump will be why.”

And he links to the following article.


The tweet is creepily prescient.

Coronavirus vs. swine flu (H1N1) – 2009-2010 Swine flu is the same strain of flu that wiped out 1.7% of the global population in 1918. It was declared a pandemic again in June 2009, and an estimated 11-21% of the world population contracted it.

Fortunately, the CFR is considerably lower than in the previous pandemic, with 0.1-0.5% of cases resulting in death. 18,500 of these deaths were laboratory-confirmed, but estimates are as high as 151,700-575,400 globally.

50-80% of severe cases were found in people who had underlying conditions such as pregnancy, asthma, diabetes, and cardiovascular disorders.

Compared to swine flu, coronavirus:

- is less widespread - has caused fewer deaths - has a higher CFR - has a longer incubation period - affects young people less


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact