The CDC’s Coronavirus Death Rate Estimates Are Too Low, Many Scientists Say | Hacker News Reader

The CDC’s Coronavirus Death Rate Estimates Are Too Low, Many Scientists Say

buzzfeednews.com

34 pointsgeorgecmu24 comments

The integrity of all new sources have been challenged to such a degree that the real loser here is the average person. There is no way for many of us to assess who to trust and we are being mired in confusion and smokescreens created by the powers that be. There is no denying that it is being done intentionally and its just sad.

It's partially the fault of the sources, but I also feel like there are unrealistic expectations. Most people I've seen discuss it online expect a lot more details with a lot more certainty than could possibly be available.

This is why half the country is willing to buy whatever the loudest voice in the room is willing to say.

I wonder how many states are also cooking the stats amidst the flurry of reopening.

It doesn't seem like there'd be much incentive to. If they have an uncontrolled outbreak, people are gonna notice regardless of what the stats say.

You really don’t see the incentive to underreport?

Just being open for business is one thing, getting workers and consumers to fulfill their roles in the system is another.

If people see the death rates are still quickly growing, they’ll be less likely to resume working and spending.

If death rates are still quickly growing, people are going to see that regardless of what the official statistics say.

I've been fortunate so far that I personally do not know a single person who has died of covid. I am aware of a family member of an acquaintance died of covid. If the death rate really spikes, more and more people will start to have someone in their personal circle who has died of covid. And people who had someone in their personal circle die will change their behavior in some way.

I agree there's a lot of incentive to reduce the counts in states where for political reasons in party they opened up their states aggressively, one example is Georgia. Georgia has somewhat surprising overall positive counts trending down. Compare them to Alabama which has been consistently going up. Georgia has reported incorrect results and Florida fired someone working on their website. https://www.cnn.com/2020/05/20/us/florida-georgia-covid-19-t...

The problem arises when people talk about a single "death rate" without the proper caveats. It is well understood that the infection fatality rate varies with age, from 0.001% for those under 20 to ~10% for those over 80.[1] When you apply this to the entire US population, you get ~0.7%.

[1] https://science.sciencemag.org/content/sci/suppl/2020/05/12/...

The local hospital around me has about 600 beds. And the highest number of patients they had was 54. Nurses hours were cut back severely since there just wasn’t much to do in the hospital. I think the us numbers are so slow because of the highish bed to person ratio. And we had plenty of warning time to free up beds and such. I’m only aware of a few hospitals that actually had major issues.

I think the problem is that tons of people get infected without bothering to get tested. It's not a simple process to get tested and if you are not very sick, you just won't bother.

In much of the country, you can’t get tested unless you have symptoms, which means you’ve been infectious for 2–14 days.

Even now, we’re still in single digits in terms of the percentage of the population that’s been tested.

From the CDC report making those "estimates"

>The parameters in the scenarios .. Are not predictions of the expected effects of COVID-19.

They are demonstrating ways of modeling the effects based only on the the available evidence of the viruses biological and epidemiological characteristics. These can later be used in more detailed models.

The observers using them to support reopening the country are misusing them, but this article is also missing the point.

The only rational option I see is Americans must be some of the healthiest people. Kidding aside this is a hard problem to solve. Some states are over reporting some are under reporting, there just isn’t much consistency. And of course there is tons of political bias so who actually knows.

Maybe Americans are just way less social than everyone thought? So there’s less viral load and less deaths.

> The only rational option I see is Americans must be some of the healthiest people. Kidding aside

The US was fortunate to have considerable ICU capacity. That has kept the mortality rate far lower than it otherwise would have been had the US been overloaded to the extent that some other nations were. Immense ICU capacity is one of the reasons Germany faired so well compared to most other nations in Europe.

By the time Italy and Spain were reporting a mere 20,000 cases, their healthcare systems were already crashing and they were forced to let older patients die because they didn't have the capacity to treat everyone. The same thing quickly happened in France and Britain as well, producing high mortality rates.

Also, most European nations are dominated in terms of population by one or two cities, as a percentage share of the overall population. You can see this in the list of primate cities [1]. So if your national population is 50 million and you have a primate city the size of NYC, and it gets hit hard by the virus, it's going to produce an extreme mortality rate nationally. With the US, you see a far more spread out context because NYC as one example doesn't represent nearly so large a share of the national population, with the US having few hyper dense cities vs its 330m population and few major cities in the climate zone that was ideal for SARS-CoV-2 to spread. If Los Angeles had NYC's climate, it would have produced a horrific outcome.

[1] https://en.wikipedia.org/wiki/Primate_city

I think we should set Italian stats aside.

It turns out that they used nursing homes to bed corona patients, so the residents were wiped out.

I'm curious why NY had such a bad time compared to SF. Was it because in a cold climate central heating circulated the corona virus droplets?

NY was afraid their hospitals would be overrun, so it sent their elderly covid patients back to the nursing homes. And then the nursing homes got wiped out. I'm not aware of this policy happening in SF.

Interesting, first I heard that NY copied Italy. I thought it was winter HVAC and overzealous ventilator use, but yeah, that would do it.

SF didn't really do anything, including testing, until very late, so we never needed more patient space, ironically. (We actually gave away 500 ventilators early on.)

I believe that's due to many Bay Area residents shuttling between SF and China, essentially starting a very early herd immunity in Dec. The Santa Clara dashboard makes corona look like a seasonal flu - there's numbers, but nothing of concern yet.

It was reported that at the end of NAMM on Jan. 24, planeloads of Chinese travellers disembarked at LAX, all wearing masks, so they knew what was up. (That was also the day Disney closed their Shanghai property due to corona.)

https://www.cnbc.com/2020/01/24/shanghai-disney-shuts-to-pre...

You would think closing a Disney theme park would catch the attention of the CDC or somebody in our government, yet here we are.

Don't believe the hype. The CDC and government were taking steps all through January. Hindsight is always 20/20, so one might suggest certain things could've been done earlier, but it's not as if coronavirus was being ignored.

Jan 3: CDC Director Redfield formally requests permission to send experts to China to investigate the new virus

Jan 5: CDC Director Redfield repeats formal request

Jan 6: CDC issued level 1 travel notice for Wuhan

Jan 7: CDC established coronavirus incident management system

Jan 11: CDC updated a Level 1 health notice for Wuhan

Jan 17: CDC began public health entry screening at 3 US airports that received the most travelers from Wuhan

Jan 20: Fauci announces NIH is working on vaccine

Jan 21: CDC activates emergency operations center to provide support for coronavirus response

Jan 23: CDC sought "special emergency authorization" from the FDA to allow states to use newly developed coronavirus test

Jan 27: Trump tweeted he made an offer to Xi Jinping to send experts to China

Jan 27: CDC issues level 3 travel health notice urging Americans to avoid all nonessential travel to China

Jan 27: White House Coronavirus Task Force had its first meeting.

Jan 31: Trump declared coronavirus a public health emergency, announced Chinese travel restrictions

... the actions just get more aggressive from here

Thanks for the info.

If those dates are correct for the first week of Jan., then the CDC was actually very aggressive for a govt. agency, since Wuhan was still trying to cover it up that week. (See link below.) We need to think about why it took 4 weeks for US action, and what a major trading economy like the US can do.

But ... note that December is missing from the timeline:

The first known Wuhan hospitalization was Dec. 1, and since I travel a lot in Asia, I learned of a highly contagious, deadly flu there on Dec. 15.

1) So it took over a month for the CDC to find out, despite US funding of the Wuhan corona laboratory (the epicenter.)

2) When talking with mainland Chinese people, they say Chinese social media had active topics about flu in Nov. and Dec. Something to search for in Chinese.

3) I looked at the CDC web site recently for how to report an outbreak, and there is no direct way for a citizen to do so. They require you to report it via an allowed channel. Again, something to think about.

"What did China know, and when?"

https://thehill.com/opinion/international/490258-what-did-ch...

In most large cities, hospital infrastructure is overbuilt by a factor of like two.

The UK’s ICUs did not get overwhelmed, even in London. A few individual hospitals came close for a few days, but on the whole the hospitals coped fine and actually spent most of the crisis well below their usual occupancy levels.

Why exactly (or even if) our mortality rate is high is still an open question.

Reply on news.ycombinator.com