Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror
×
Medicine United States

The CDC Says Its New 'Best Estimate' Is That 0.4 Percent of People With Symptoms and COVID-19 Will Die (cnn.com) 334

An anonymous reader quotes a report from CNN: In new guidance for mathematical modelers and public health officials, the U.S. Centers for Disease Control and Prevention is estimating that about a third of coronavirus infections are asymptomatic. The CDC also says its "best estimate" is that 0.4% of people who show symptoms and have Covid-19 will die, and the agency estimates that 40% of coronavirus transmission is occurring before people feel sick.

The numbers are part of five planning scenarios that "are being used by mathematical modelers throughout the federal government," according to the CDC. Four of those scenarios represent "the lower and upper bounds of disease severity and viral transmissibility." The fifth scenario is the CDC's "current best estimate about viral transmission and disease severity in the United States." In that scenario, the agency described its estimate that 0.4% of people who feel sick with Covid-19 will die. For people age 65 and older, the CDC puts that number at 1.3%. For people 49 and under, the agency estimated that 0.05% of symptomatic people will die.
The CDC cautions that these numbers are based on real data collected by the agency before April 29 and are subject to change. Still, the "current best estimate" number of 0.4% is significantly lower than the 3.4% mortality rate the World Health Organization warned in early March.

"As I see it, the 'best estimate' is extremely optimistic, and the 'worst case' scenario is fairly optimistic even as a best estimate. One certainly wants to consider worse scenarios," biologist Carl Bergstrom of the University of Washington told CNN. "By introducing these as the official parameter sets for modeling efforts, CDC is influencing the models produced by federal agencies, but also the broader scientific discourse because there will be some pressure to use the CDC standard parameter sets in modeling papers going forward," he said. "Given that these parameter sets underestimate fatality by a substantial margin compared to current scientific consensus, this is deeply problematic."

UPDATE (5/28): An epidemiologist at the Johns Hopkins Center for Health Security also contradict the CDC's low numbers, saying that after reviewing new widespread testing, "The current best estimates for the infection fatality risk are between 0.5% and 1%," according to NPR.

"But even a virus with a fatality rate less than 1% presents a formidable threat, Rivers says."

Further reading: Many Scientists Warn CDC's New Death Rate Estimates Far Too Low.
This discussion has been archived. No new comments can be posted.

The CDC Says Its New 'Best Estimate' Is That 0.4 Percent of People With Symptoms and COVID-19 Will Die

Comments Filter:
  • by ardmhacha ( 192482 ) on Wednesday May 27, 2020 @08:17AM (#60109612)

    There have been at least 20,000 Covid deaths in NYC. A 0.4% death rate means 5 million symptomatic cases, A one third asymptomatic rate means 7.5 million total cases

    The population of NYC is just over 8 million.

    • by matthewmok ( 412065 ) on Wednesday May 27, 2020 @08:28AM (#60109666)

      There have been at least 20,000 Covid deaths in NYC. A 0.4% death rate means 5 million symptomatic cases, A one third asymptomatic rate means 7.5 million total cases

      The population of NYC is just over 8 million.

      And 10,000 or more of those were from nursing homes and those were because they sent Covid-19 positive patients there....when they KNEW that the sick/elderly were the most vulnerable.They did the same thing in England.

    • by Rockoon ( 1252108 ) on Wednesday May 27, 2020 @08:29AM (#60109670)
      Except NY made sure the most vulnerable population got it, by (A) sending covid patients into their homes, and (B) not informing the homes that they were receiving covid patients

      You cant go by a single NY stat, because they are killing people on purpose.
    • by hattig ( 47930 )

      This is why a large-scale antibody test has to be done, to find out to a reasonable confidence just what percentage of a population actually has had this.

      That that shows that, e.g., 20% of the NYC population had C19, then we can see that the death rate would be far higher (~1.7%).

      But if 80% have had it and the area is now 'herd immune' then this would be excellent news (for NYC).

      • by Nidi62 ( 1525137 )

        This is why a large-scale antibody test has to be done, to find out to a reasonable confidence just what percentage of a population actually has had this.

        That that shows that, e.g., 20% of the NYC population had C19, then we can see that the death rate would be far higher (~1.7%).

        But if 80% have had it and the area is now 'herd immune' then this would be excellent news (for NYC).

        Except the CDC says that current antibody tests are roughly 50% accurate, with a high rate of false positives (moreso than false negatives). There's both no solid indication of post-infection immunity and it is my understanding that a lot of the antibody tests will show positive after other coronavirus infections, not just covid19. They do not recommend making policy decisions off antibody tests.

      • by Zocalo ( 252965 )
        It is, but of those that have been done the figures are all over the map - often because the population has been "enclosed", e.g. on a cruise liner, aircraft carrier, or in a prison where the demographics are also going to be heavily skewed. Establishing this percentage used to be a key point of the UK's strategy that seems to have been sidelined a bit of late. Since the ONS is currently undertaking an on-going study on current infections [ons.gov.uk] I suspect that's because a suitably reliable anti-body test that ca
    • Comment removed based on user account deletion
    • The 0.4% death rate is for the country overall, the vast majority of whose hospitals are not at capacity so every Covid-19 patient can receive the best treatment available.

      NYC allowed the virus to get ahead of them, and their hospitals got overwhelmed. People didn't always get timely treatment, or any treatment. That leads to a higher fatality rate.
    • There have been at least 20,000 Covid deaths in NYC. A 0.4% death rate means 5 million symptomatic cases, A one third asymptomatic rate means 7.5 million total cases

      The population of NYC is just over 8 million.

      My thought exactly.

      Last I heard the best estimate for NYC is that 25% of the population was infected.

      So that gives a 20,000 / (8,000,000 * .25) *100 = 1% infected, not just symptomatic death rate.

      I don't want to Dunning Kruger myself, but that looks like a discrepancy that needs to be addressed.

  • by CastrTroy ( 595695 ) on Wednesday May 27, 2020 @08:17AM (#60109616)

    So, if 0.4% of people die, and New York state has had 30,000 deaths, then that means 1 in 26 people in New York State has had Coronavirus and its symptoms? And if the numbers for asymptomatic cases are between 25 and 50 percent, that means that somewhere around 1 in 13, to 1 in 20 people have already had the disease.

    • Actualy, it's too early, and I didn't to the math right, and slashdot doesn't allow editing. I was off by a factor of 10 and used .04 instead of .004 for 4%. This means that 7.5 million New Yorkers have already had symptoms and the virus, and are now at a 37.5% infection rate.

      • This is roughly in line with antibody testing that has shown that about 20% of New Yorkers have antibodies to COVID.

        https://patch.com/new-york/new... [patch.com]

      • Umm, 4% = 4/100 = .04. Need another cup of coffee? Using your numbers:
        30,000 deaths / 4% = 750,000 people with symptoms
        750,000/(1-25%) = 1 million total infections
        750,000/(1-50%) = 1.5 million total infections
        New York state population = ~19 million

        So by your numbers somewhere between 1 in 13 and 1 in 19 people have caught it so far, and you were right the first time

        No idea where you got the 37.5% infection rate though: 1/13 = 7.7%, 1/19 = 5.2%

    • A lot of those deaths were preventable. The people died because New York screwed up their initial response, and allowed their hospital system to become overhwlemed. They couldn't provide all patients with timely treatment, so the fatality rate there was higher.

      That's the whole point of flattening the curve. Prevent your hospitals from being overwhelmed so the fatality rate doesn't unnecessarily increase. Most of the country has successfully flattened the curve sufficiently, and the 0.4% fatality rate re
  • Seems reasonable (Score:5, Insightful)

    by The1stImmortal ( 1990110 ) on Wednesday May 27, 2020 @08:18AM (#60109620)

    The numbers aren't too far off what Australia is seeing. About half the known cases here have been 49+, and our fatality rate has been hovering around 1.5%. There's known to be cases in the community we haven't identified, so that will push the effective number down further. Given the number of unknowns involved still, particularly in the US, I'd say the CDC numbers are reasonable ballpark figures.

    It's very odd that the article quotes (and the submission above doesn't make it clear where the quote comes from) a computer modeller/simulator as doubting the numbers. In this kind of situation, the modeller isn't the one getting the primary data, they're given it by people closer to the ground. It's their job to build simulations of outcomes based on the facts as they're given them, not to try to second-guess the source material.

    A quote from an epidemiologist or something would make more sense.

  • "the agency estimates that 40% of coronavirus transmission is occurring before people feel sick."
    So 60% went out in public with a cough and fever? I DOUBT IT.
    Also, 0.4%? Even if you assume half the recorded worldwide deaths are incorrectly attributed, that's 44 million infected. So the 5.7 mil known cases compared to that many unknown actually sounds about right until you consider all the people that allegedly carry it without ever having symptoms. Like they get an innoculation dose but somehow remain con
    • by stdarg ( 456557 )

      So 60% went out in public with a cough and fever? I DOUBT IT.

      People who live with others can spread it without going out in public. Remember, our brilliant health officials told everyone, if your symptoms are mild then stay home, you don't need a test. So this probably covers most cases. People spreading it around to their families, people in their apartment buildings, neighbors, etc.

      People who don't live with others, especially other adults, probably did have to go out, they still need food and stuff. Many people either can't afford (poor people) or don't know how t

    • by Rei ( 128717 ) on Wednesday May 27, 2020 @09:05AM (#60109840) Homepage

      So 60% went out in public with a cough and fever? I DOUBT IT.

      1. The highest rates of transmission are within the household.
      2. People go out with minor symptoms (and sometimes with major symptoms) all the bloody time. Diseases rely on humans doing this.

  • by Joe2020 ( 6760092 ) on Wednesday May 27, 2020 @08:24AM (#60109650)

    At the start of the outbreak did people get scared, even panicked, they felt compassion for one another, help where they could and came up with ideas on what to do.

    But once things got under control and most were safe do people turn it into a competition. Suddenly it's "my numbers are better than your numbers", "our model is the best model" and "my hindsight is sharper than yours".

    Are we just bored and waiting for the next big epidemic to hit us?

    • All I can say is that we can be glad that this disease really wasn't worse. We failed the drill. When the real big one comes, we know it'll all be over.

      • Yup, you can never improve off the first drill!

        That's why there are more drills, just to confirm you're still doing the same thing as that first drill.

  • Are there any numbers yet detailing the risk of permanent lung damage after surviving the virus? Or is it still considered anecdotal rather than a pattern?

    • by Merk42 ( 1906718 )
      I'm not a medical professional, but, if the virus has only been known about for a few months, how would they know if anything (outside of death) is permanent?
      • by Calydor ( 739835 )

        For one thing they could look at the physical damage and see if it's a kind that's known to be naturally healed over time (eg., like a cut on your arm) or be permanent (eg., your arm got cut off).

      • There are degrees of scarring and damage that can be determined to be permanent, even if you haven't had years to study how the patient recovery progresses. My sister is head of anesthesiology at the VA, and she is seeing cases with permanent lung and organ damage. Her numbers are small - she's seen something like 300 cases - and this damage doesn't strike every patient, or even every patient who spends time on a ventilator. It happens she says in a higher percentage of ECMO patients. So we're still som
    • by Rei ( 128717 )

      Lung damage is what pneumonia frequently does. If you develop pneumonia (esp. if you need invasive ventilation), regardless of the cause, you're likely to end up with some reduced lung capacity. If you don't, you're not.

      It's not COVID-19 specific.

  • by nysus ( 162232 ) on Wednesday May 27, 2020 @08:31AM (#60109678)

    It's a real shame when you don't know how much of the science has been politicized.

    • It's worse than that, we're looking at best guesses.

      The difference between a guess and an estimate is the amount of hard data backing it up. Unfortunately, we don't have any accurate figures on

      a) How many people are currently infected. Nice to know, but not relevant for mortality rates
      b) How many people have been infected. Until we get a reliable antibody test and a large randomised sample we have no real numbers.
      c) How many people have actually died of Covid-19. That is excess mortality above what would

  • If only .4% of those showing symptoms will die, and we are stating that in the USA 100,000 have died, then we should have 100000/.004 or 25 million people showing symptoms. Since I have already heard that state agencies are trying to suppress deaths, and I have seen images of bodies loaded in trucks and stories from health care professionals stating the death rates I tend to believe the 100,000 figure is at the low end of any real number.

    However I don't see 25 million people exhibiting symptoms....

    • then we should have 100000/.004 or 25 million people showing symptoms.

      We already know that in some the symptoms are no worse than a standard cold. These people were never going to get tested - either because of no availability of tests or because of not thinking they're sick beyond rhinovirus or a milder coronavirus. Although I'm still surprised that they are estimating on "symptomatic cases" when there are so many asymptomatic cases.

      Just how many is going to be a mystery and widespread antibody testing will only paint a partial picture.

    • I tend to believe the 100,000 figure is at the low end of any real number.

      Since February 1st, Kentucky has had less than 400 COVID-19 deaths. However, they've had more than 1,000 more pnemonia deaths than in 2019 during the same period. And the social distancing should have helped reduce the number of pneumonia deaths due to other causes.

      Just to give you an idea of how underreported it's been in Kentucky alone.

    • Actually, the number of people who have had the virus is likely higher than 25 million. A New York antibody study showed that 20% of New Yorkers already had antibodies for the virus.

      https://patch.com/new-york/new... [patch.com]

      Across the US, 20% of 300 million would be more like 60 million.

      So where are all these people in the numbers? Well of course, people with mild symptoms aren't going to hospitals, and are largely uncounted. My wife had it in March, but never got tested. I'm sure I got it too, because I certainly d

  • During a recent Whitehouse press conference, Dr Fauci indicated that as many as 50% of those who become infected never become symptomatic. So that puts the rate as low as 0.2%.

    Given that the Dems are "science driven" I'm sure blue state governors will review this new information and immediately open their states. I'm looking at you Newsome, Cuomo, and Heir Whitmer.

    This entire lockdown has been a farce, an test to see how easy it is for government to incarcerate the entire population in their own homes and

    • Given that the Dems are "science driven" I'm sure blue state governors will review this new information and immediately open their states.

      Science-driven approaches are not quite so reactionary. This is one data point / estimate among many. Gradual opening, while spacing out stages of at least 14-day increments to monitor actual infection rates is exactly what's planned for Illinois. Bumping up the timeline by 4 or 5 days on just one report is not science-based. That's just following the results you want to see.

      You can use hindsight all you want to blame this person or that for overreacting, but you're doing this with more information than

    • then let those who are in fear stay home while the rest continue with the business of living our lives.

      For the same reason that we don't declare be "law-free purge day", where those in fear can stay home and the rest can continue with the business of living our lives. Hopefully the reason is self-evident.

  • My state's published numbers show a mortality rate of confirmed cases at about 3.7% (since people don't drop dead immediately after being tested, the death rate is more likely to go up than down, but let's just use that). Since testing has mostly been limited to people with some symptoms already, that would mean that 90% of the people WITH SYMPTOMS haven't even been tested? I know the testing rate is not great (since test supplies have been low), but I find it hard to believe it is THAT low.

  • by slack_justyb ( 862874 ) on Wednesday May 27, 2020 @09:06AM (#60109848)

    From the summary above.

    The CDC cautions that these numbers are based on real data collected by the agency before April 29 and are subject to change. Still, the "current best estimate" number of 0.4% is significantly lower than the 3.4% mortality rate the World Health Organization warned in early March.

    From the actual CDC report

    Symptomatic Case Hospitalization Ratio, stratified by age in years

    0–49: 0.017 | 50–64: 0.045 | 65+: 0.074 | Overall: 0.034

    Which indicates a 3.4% mortality rate which is still very much in line with the WHO estimates. What's important to gain from these numbers is this really important highlight.

    Symptomatic Case Hospitalization Ratio

    What this is saying is that if you only focus on those who present themselves to the hospital, this is your mortality rate. Now the other number which is...

    Symptomatic Case Fatality Ratio, stratified by age in years

    0-49: 0.0005 | 50-64: 0.002 | 65+: 0.013 | Overall: 0.004

    This is those who showed symptoms and presented or did not present themselves to the hospital who ultimately died. The latter being a self-quarantine situation as an example. So those that stayed home and got better bring this number down. Of course, WHO numbers would have only been the first category, how would the WHO have numbers for a testing they don't run. Now that's still nothing to sneeze at, because those rates are still really bad compared to flu. And yes, the numbers for flu deaths are deaths reported so that first number, not the second one, we don't typically gather data for who got flu and just got better on their own. In fact, we usually don't gather that data for any disease. If you don't go to the hospital/medical office/doctors office/that place in your grocery store..., statistically speaking, you're just 100% healthy.

    So there's a bit of context that might be required for the 0.4% value there. Because the CDC numbers are still in sharp agreement with the WHO estimate. The 0.4% is measuring something completely different and it's really important that people understand that slight difference. And I only say all of this because the summary does the thing that's wrong here. It compares the CDC case fatality number to the hospitalization fatality number and those are two slightly different things. It's not apples to apples saying that 0.4% is less than 3.4% since those two numbers aren't measuring the same thing. There's a lot more people we don't usually count in the that 0.4% and that context is pretty important.

  • CFR = 0.4% implies IFR of 0.26%. If it takes 60% infected to get herd immunity (some say higher; I've heard 80%), and if that infection is spread evenly across age groups, then the U.S. can expect another ~410k deaths absent the development of a vaccine or effective treatment.
  • by NotSoHeavyD3 ( 1400425 ) on Wednesday May 27, 2020 @09:27AM (#60109970) Journal
    Since it's definitely more if you're older. Also the gist of the USS Teddy Roosevelt is about 1200 got it, 40% had symptoms and 1 died.(Or 0.2% of symptomatic cases ended in death.)

Technology is dominated by those who manage what they do not understand.

Working...