Skip to content

The best coronavirus summary so far

I’d still go with this article by Ed Yong, which covers biology, epidemiology, medicine, and politics. Here’s one bit:

In 2018, when writing about whether the U.S. was ready for the next pandemic, I [Yong] noted that the country was trapped in a cycle of panic and neglect. It rises to meet each new disease, but then settles into complacency once the threat is over. With COVID-19, I fear that the U.S. might enter the neglect phase before the panic part is even finished. . . . The virus is disproportionately killing people in low-income jobs who don’t have the privilege of working from home, but who will nonetheless be shamed for not distancing themselves. . . . If that happens, the panic-neglect cycle will inevitably continue. The U.S. will miss the chance to reexamine how systemic failures of its health-care system left so many citizens vulnerable, or to put in place measures that might forestall another resurgence. It will get hammered by the same damn virus again, and be driven into more severe lockdowns. . . .

There’s also lots of details on testing and measurement, the health-care system, and possible decision points.


  1. Yong is a treasure. I’ve been considering subscribing just because the articles he writes are so good.

  2. D Kane says:

    > Yong is a treasure.

    No, Yong is a hack. Or, at least, I have clear (to me) evidence of one bit of hackery. But the problem, with Yong, is not, mainly, pimping out a garbage PNAS observational study. We all make mistakes. The problem is that Yong had zero interest in actually talking to a knowledgeable critic, i.e., me. You think, for this article, he talked with anyone opposed to his point of view? I doubt it!

    • somebody says:

      I agree that the female doctors article is garbage, but it’s a pretty flat report of what the authors are saying. If uncritically reporting that someone made a claim at least once, then there no one isn’t a hack.

      > You think, for this article, he talked with anyone opposed to his point of view? I doubt it!

      This is a completely pointless statement. Either give a specific criticism of the story where consulting with someone who disagrees would have improved the content, or don’t say it at all. Criticizing the research methodology that you imagine someone probably followed without any link to ink on the page is entirely pointless; it’s equivalent to saying “i dislike the article because i personally feel like the research process was probably bad”.

      • somebody else says:

        But you don’t understand! Yong had the gaul not to consult with Kane before he published his article. He should have known that Kane is the gatekeeper for dissemination of peer-reviewed scientific results! Don’t we all!

    • Anonymous says:

      You should take yourself less seriously.

  3. Zad says:

    You’ve shared this quite a few times, you must really like this one!

  4. Dr. Eric Wornhoff says:

    Is it really a “privilege” to be able to work from home?
    It sounds unfair (or at least, unequal). Maybe these “privileged” people should have their “privileges” revoked, perhaps by a Worker’s Tribunal”, or “People’s Council” or something appropriate like that.

    • Andrew says:


      I don’t know if it’s a privilege to work from home, but I guess it’s a bit of luck to be able to still get paid for working, at a time when so many people are losing their jobs. I assume that the hypothetical tribunal would be more interested in getting their own jobs back than in stopping other people from working. For example, if any of the tribunal have kids in school, they’d probably like to still allow their kids’ teachers to be able to work from home.

  5. Dr. Eric Wornhoff says:

    OK, I get it now. “Privileged” means “lucky and worked hard (enough to get what they got)”. Sounded like it meant that someone gave someone something they didn’t necessarily deserve (at least, no more than someone else might have deserved it) and that it could and probably should be taken away and given to someone else. Somehow to my ear when I hear the word privileged I expect to hear the words white, discrimination, and racism. But as you explained it just means mostly “lucky”.

    • Martha (Smith) says:


      I appreciate your reading Andrew’s comment carefully, realizing that you and he had different default interpretations of “privilege”, and responding as you did.

      But I would also like to point out that there is a connection between the two meanings of “privileged” — namely, that (regrettably) there are categorizations of people in which some people, merely because they belong to some particular category (e.g., “white”) , are (given current realities) more likely to be lucky in some ways (thus are more privileged) than people in some other categories (e.g., “black”).

  6. Terry says:

    “The virus is disproportionately killing people in low-income jobs…. The virus is disproportionately killing black people…. The virus is disproportionately killing elderly people….”

    The virus is also disproportionately killing men. I wonder why that was left out.

  7. Anonymous says:

    The U.S. will miss the chance to reexamine how systemic failures of its health-care system left so many citizens vulnerable

    Why do people feel the need to make this obviously wrong connection? Like did this guy even think about this a tiny bit? China and Italy have universal healthcare and had a worse problem. If anything the correlation goes the other way.

    • Rob Jonsson says:

      His politics got in the way of writing an honest article.

      • David J. Littleboy says:

        Given it’s now the US with the worse problem, maybe he was right after all.

        But your politics seem to be getting in the way of noticing that…

        • DAVID SAMRA says:

          False – USA has far lower cases per capita than any universal health care country.

          But don’t let facts get in the way of your ideology.

          • Curious says:

            Testing, testing, 1, 2, 3. Wanna keep your cases low? Don’t test. Pretend it’s not happening. If you don’t test post-mortem, then the deaths cannot be attributed to the crisis you bungled either. As a way to hide things, not a bad strategy. As a way to govern competently — it’s a bit of a problem. And it will all come out at some point.

          • Andrew says:

            David Samra:

            You write, “USA has far lower cases per capita than any universal health care country.” This seems like a false claim. Here’s a list of countries with universal health care. Here’s a list of countries with coronavirus cases per capita. There are lots of countries with universal health care that have rates lower than the U.S.

    • Responder says:

      Notice the sentence you highlighted doesn’t say anything about: (1) other countries (2) universal healthcare.
      One premise of the statement seems to be that `US health-care system left so many citizens vulnerable’ — if we look at the number of deaths, this seems to be true.
      Second premise of that statement is that systemic failures in health-care system led to this. This is un-substantiated.
      Third premise is that US will miss the chance to re-examine. This part needs context.

  8. Bill Spight says:

    I don’t think that the neglect part of the panic-neglect cycle in the US is complacency. I think it is deeper than that. I think that it is part of the erosion in American culture of the related virtues of prudence and preparedness. We are prepared, OC, for WWIII. The Boy Scout motto is Be Prepared. That motto fit our culture in the early 20th century, but prudence and preparedness are costly, and require taking a long term view. Preparing for risks that don’t materialize within five years results in charges of waste and mismanagement. That President Trump closed the pandemic preparedness office is hardly surprising. Not only had Obama created it, but someone who has gone bankrupt as often as Trump cannot be expected to exercise prudence.

    Since this is a statistics blog, let me note that, curiously, the Kelley criterion for making bets not only guards against going bust, it also compounds the expected return on investment. Prudence can be profitable.

  9. Jonathan says:

    Everyone has an opinion now, and all likely are wrong in material ways. I include my own beliefs. Example, at least for me: I live in Boston and the State publishes each day a list of deaths (by age in decades, by yes or no for pre-existing conditions) and total confirmed cases (by decades), so I keep a list. Guess how many people have died under age 50 up to reporting 4/18? 16 total, 6 in their 30’s and 10 in their 40’s. That’s out of 1560 deaths. All but one are yes for pre-existing conditions, with the remaining one as unknown. That’s almost exactly 1%. Guess how many people in their 100’s have died? 25. For context, there are about 1500 centarians in the state at any given time (so similar to the number of Covid-19 deaths recorded here so far), so about 1.5% of the entire centarians population. I didn’t check for pre-existing conditions for centarians because, let’s face it, being 100 is a pre-existing condition. Note that not a single person below age 30 is recorded. And all but one below age 50 are yes for pre-existing conditions.

    FYI, you can check the information here: – updated at 4PM every day.

    If you go through the confirmed cases, the number of people in their 20’s confirmed is about the same as people in other decades, higher than in the 70’s, a peak in the 50’s, but relatively even allowing for who shows up to get tested, the age of the healthcare worker population, etc.

    Rather obviously, if we assume more people are infected, the death rate for people under age 50 in MA is not much, and disappears for people below age 30, especially if you take out whatever they mean by pre-existing conditions. This says to me the risk of death for younger people is much lower than we currently want to admit. I said ‘want to admit’ because the mantra now is ‘we’re all in this together’ and the media highlights every young person who either dies or who has recovered after a battle. I have trouble believing no one in the Boston media has managed to take the 5 minutes it took me to go through the lists, especially now that they sort by decade. My guess is they don’t want to say this because that would encourage young people to go out and to demand they can go out. Maybe I’m wrong but the work is so obvious and easy that I must ask the question.

    I’m sorry if this is getting long but the data from MA says that we should think about younger people differently, and that we should think about a set of pre-existing conditions among people under say age 50 (and I note there are relatively few deaths in the 50’s, but the risk rises by a few fold). This is very different from the linked article. Example: when Covid-19 appeared, my belief was that we should quarantine all elder care facilities immediately. Rather than do that, the state sent in the National Guard – in hazmat suits – to test people. According to the Globe, this drove away staff that they’ve been completely unable to replace, and that has led to – by far – the largest number of infections and deaths occurring within facilities which should have been locked down but weren’t. (See this as an example: Our governor has been applauded for running terrific daily press briefings but the actual actions of the state appear to have failed with respect to the literal most vulnerable. Isn’t that the kind of lesson we should learn? (As an aside, they tested in nursing homes as though that was a preventative, as though that one time reading was sufficient. Really bad.)

    There’s been a tremendous amount of hysteria. Your governor 2 weeks ago was almost crying ‘you tell me which 26,000 people will die without ventilators’ and all my friends wanted to draft him to replace Biden, but now it appears ventilators aren’t necessarily a good idea. So people jumped up and down applauding one hysterical reaction while denigrating others. The national news (we mostly watch CBS) denigrates Trump’s favorite drug while running reports on drugs that are as speculative or worse, seemingly without self-awareness.

    Another example: there’s a lot of really incomplete research coming out and that gets hyped as though it’s gold. One issue is ventilators and the growing sense, but another is the concept of viral load at exposure. That matters in other diseases and some ‘current’ research indicates viral load directly correlates to the immune system crash that endangers people, and which particularly may expose healthcare workers. (See this article for both sides of that emerging issue: ). This suggests that what matters is the characteristics of the people exposed. That is where public health requirements have collided with the actual but as yet undetermined ‘statistics’: everyone must comply because we aren’t willing to think about separating people by risk level. That may be over-stated, but it suggests we may be crippling our economic life out of fear of ‘infection’ rather than what ‘infection’ means for the people infected. That connects to the magical thinking people have about how the economy works: as in, restaurants and other businesses can open at some much lower capacity … except then you’re saying they need to operate at a loss, need to order supplies without cash flow, etc., when we can’t even manage to get loans to people so they can pay their employees. (This is the Globe reprint of the NYT story: .)

    Finally, regarding inequities, if you look at the rest of the world, they have terrible information about how their minorities fare, so it looks like the US system fails because it doesn’t cover people. The UK’s data – which was put together by journalists because the government doesn’t release the numbers – suggests that under NHS black people are dying at a higher rate compared to population percentage than in the US. Same with South Asians. So when I read about how we can learn about inequities, my first reaction is something I learned from my father. He turned on the radio and ran to his dad to tell him Martians had invaded the earth. My grandfather thought for a second and asked, ‘Is it on every station?’ It seems that inequity along minority/poverty lines occurs with similar frequency across healthcare systems. (That is, based on what I read about other countries, which is a mess of bad reporting, plus a few reasonably reliable reports.) So if inequity is on every station, then it approaches absurdity to argue that somehow we can learn how to fix the problem based on our experience. That becomes a political argument in favor of a different healthcare system – which, btw, I favor – based on what is happening on one radio station.

    • Brent Hutto says:


      Your comments fairly closely match my own beliefs about the situation. For whatever that’s worth.

    • Jonathan (another one) says:

      Of the 1560 MA deaths, 810 have occurred in nursing homes. In my small town in far northwestern MA, the town figure of 74 infections caused huge alarm until officials hastened to add that 45 of them are in a single nursing home in town.

      The original British plan was to “let it rip” while full-out sequestering those over 60 and those under with pre-existing conditions. I’m still not sure why that wouldn’t have been a better strategy than flattening the curve, which extends the period of virus spread without, so far as I can changing the death toll very much at all, at least so long as the hospital admission rate stays below capacity, which it has. I guess Italy is the counter-example, but the lack of sequestration there (and indeed, the impossibility given the fact that aged relatives are far more likely to live with their families) makes it non-comparable.

    • Dale Lehman says:

      This certainly casts a different light on things. But have you looked at hospitalizations? I don’t know if the data is available by age, but before we jump to conclusions about age warfare, it might be important to understand how many younger people might be getting very sick but not dying. And, if they are a sizable group, then does their survival depend on having access to medical care? So, maybe we’re not really all in this together, but perhaps we are.

      The issue of preexisting conditions is also worrisome. Lots of people worry about the huge proportion of national income that is spend on health care, and we know that the distribution of health care costs is highly skewed. So, the health care cost burden in the US is definitely not equally shared. When considering national health care policy, would you say we are all in this together? If not, then I am not sure I like where this is headed. On the other hand, I would agree that we shouldn’t pretend that these burdens and risks are equally shared by all.

      • Carlos Ungil says:

        In Spain the age distribution for cumulative intensive care admissions is broadly one third over 70, one third between 60 and 70, and one third under 60. There distribution used to be a bit more skewed towards the elderly, I can think of two different reasons that may contribute to the change: younger people take longer to develop more severe symptoms and the eligibility for admission has changed to favour the young.

    • Martha (Smith) says:

      Jonathan said,
      “This says to me the risk of death for younger people is much lower than we currently want to admit. I said ‘want to admit’ because the mantra now is ‘we’re all in this together’ and the media highlights every young person who either dies or who has recovered after a battle.”

      My understanding of “we’re all in this together” is that one individual’s actions can affect others’ risk of contracting the virus. In particular (if I am not mistaken) there seems to be pretty strong evidence that people can contract the virus but be asymptomatic, yet still able to transmit the virus to others. So one factor in reducing the spread of the virus is for everyone (including those who are asymptomatic) to practice social distancing, out of respect for the health of others. For example, I don’t know whether or not I have the virus (I haven’t been tested for it), and I don’t have any symptoms of it, but I have started wearing a face mask when I go grocery shopping, more to do what I can to help stop the spread of the virus, than to protect myself from it. That’s what “we’re all in this together” means to me.

  10. Jonathan says:

    Realized I forgot to note the State only started breaking out deaths by age in late March, so there can be a few younger deaths ‘missed’. The total number of deaths by then was under 40.

    This was meant to reply to a comment that hasn’t appeared.

  11. Nick Adams says:

    Synthesis of data from all countries (at various stages of pandemic and with varying testing and mitigation policies):
    The IFR (infection fatality rate) is about 0.1%. This is much less than the CFR (case fatality rate) because of the high proportion of asymptomatic/mildly symptomatic cases.
    99% of deaths occur in those with a co-morbidity ( hypertension/diabetes/cardiac failure). These co-morbidities are correlated with age, obesity and male sex, hence these groups are disproportionately represented in the mortality data.
    The clinical disease is a typical viral pneumonia which when severe is associated with multiple organ failure.
    Overall, there is nothing special about this Coronavirus apart from the fact that it is novel and no-one is immune. A number of coronaviruses circulate through the community to which most people have immunity. They regularly kill old people (pneumonia is the commonest cause of death in those with dementia). We rarely see severe viral pneumonia in young people but it does occur. Because most of the population is immune, these deaths occur very infrequently and nobody really notices. When no-one is immune we have a pandemic and everyone notices.
    Anyway, I am pleased that the IFR for the young and healthy is 1/100 000. Isolating everyone else who is at risk can reduce deaths but has other consequences as we have seen.
    (All the above figures have a degree of uncertainty about them of course, but I think they are in the ballpark).

    • Nick Adams says:

      Forgot to say that the IFR rate applies only if appropriate medical care is provided. The rate will be different with an overloaded medical system and outside the first world.

    • Brent Hutto says:

      Your estimate of 0.1% IFR (which I think of as “one in a thousand”) would imply that when a country, region or perhaps state reaches around 60% to 80% infection prevalence then about 0.06% to 0.08% (60 to 80 per 100K) will be the end-stage portion of the population who have died.

      New York City and the Lombardy region (perhaps other regions I haven’t noticed) are both beyond that 60-80 per 100K rate while New York state and the countries of Italy and Spain are near or beyond it, depending on how much undercounting you posit.

      So I suspect you 0.1% can’t be quite high enough to explain the very highest mortality areas. Although that number isn’t as simple to interpret as it might seem. When we bandy about a concept like “The Infection Fatality Rate” for a disease that is so selective in targeting certain sub-populations then it necessarily implies a certain demographic profile of the area in question.

      To take an extreme example, the Infection Fatality Rate for a certain nursing home won’t be “one in a thousand”. Probably more like “one in ten” or “one in four”, depending on just how old and comorbid that home’s residents are. And in an extremely young and healthy population, let’s say members of a college sports team or recruits at an Army basic training facility, the IFR might be so low as to be almost impossible to distinguish to zero.

      Which is a long-winded way of saying your 0.1% IFR might be close to correct for some countries or states. But at least in Lombardy and NYC it implies nearly 100% infection had happened by a couple weeks ago, allowing for delay from infection to death. So overall in USA and Europe I’m guessing you are off by a factor of maybe 2 to 10.

      • Nick Adams says:

        True, it could be optimistic. But patients in Lombardy and NYC have not received optimal care because the health system was over-whelmed. The current CFR in Australia is 1% with no overload of ER/ICU. It is not unreasonable to presume that only 10% of cases have been ascertained as our testing regime has been really quite limited, hence IFR of 0.1%.
        No new community acquired cases in the state of Victoria last 48 hrs (pop. 6 million). We had the advantage of forewarning from other countries examples and locked down early and hard. The federal and state governments were in agreement and co-operated. That is what a federation of states is meant to do – work together. I feel sorry for all the US citizens who have died and will die because of…I don’t know what.

        • Carlos Ungil says:

          > Overall, there is nothing special about this Coronavirus apart from the fact that it is novel and no-one is immune.

          Maybe you could add to the list of special things about this coronavirus the fact that when it spreads (something that happens easily, as no-one is immune) the CFR is no longer 0.1% (assuming it’s as low as that in an ideal situation).

          • Yes, and that when it hits NYC rather than a couple million people getting a runny nose, minor cough, and maybe staying in bed for several days, the hospitals collapse and they dig mass graves on Hart Island and fill them as fast as they can.

          • Zhou Fang says:

            I’d suspect that under “optimal care” the CFR for these other diseases would be far less than 0.1%.

            • Carlos Ungil says:

              “Optimal” may mean anything from non-resource-constrained standard of care after the onset of clinical symptoms to early detection and the use of experimental treatments (but the right ones, or it wouldn’t be “optimal”).

              In any case, as it has already been mentioned (maybe in another blog entry) defining a CFR requires to define an universe of cases. If we think of the whole population Australia is not quite like Italy. There are 30% of people over 60 in Italy, only 20% in Australia (and the difference is even more noticeable in the tail). Given the big differences in individual CFR with age the age-weighted CFR may change a lot.

              Now that I think of it, as the infection spreads I would expect the CFR to go up even if we ignore capacity issues because the mix of infected people changes. Those importing the virus are not the more fragile people, but as local transmission increases the distribution of infected people will become closer to the true distribution of the population.

    • Martha (Smith) says:

      Nick said, “A number of coronaviruses circulate through the community to which most people have immunity. They regularly kill old people (pneumonia is the commonest cause of death in those with dementia). We rarely see severe viral pneumonia in young people but it does occur.”

      My understanding is that the high rate of death from pneumonia in people with dementia involves several factors:

      People with dementia are more likely than average to have compromised immune systems.

      People with dementia are more likely than average to be in a group care or hospital environment, where “hospital acquired pneumonia’ (which is usually antibiotic resistant, hence more likely to lead to death) is more likely to occur.

      People with dementia often have “dysphagia” (difficulty swallowing), which allows food to be aspirated into the lungs rather than going down the esophagus, which causes “bronchopneumonia” — which itself is one of the leading causes of death in people with dementia

      People with dementia are not able to cooperate with treatment for pneumonia (as well as with other medical treatments)

      • Nick Adams says:

        All true, Martha.
        The immune system changes with ageing are fascinating. When we are born we have not been exposed to any infections and only have antibodies transferred from our mothers (which don’t last). Too compensate for this, children have a very active innate immune system which responds non-specifically to foreign (microbial) proteins. As we age we are exposed to the infectious diseases circulating in our community to which we develop antibodies (very specific, focussed immunity which is far more efficient). When we reach adulthood, the innate immune system begins to be turned down because it is no longer needed as much (and consumes a lot of energy). This age-related alteration in immune function works fine until some novel pathogen appears, to which no adults have immunity.

    • jim says:

      What would you say to the idea of isolating everyone over age 60? Off the top of my head I have no idea how much sense that makes. But from what I’ve seen it would eliminate ~80% of fatalities.

      Just curious.

      • Zhou Fang says:

        What do you do about the carers, the food deliverers, the cleaners, the doctors, family members…

        • Martha (Smith) says:

          Good question.

        • jim says:

          “What do you do about the carers, the food deliverers, the cleaners, the doctors, family members…”

          • jim says:

            “What do you do about the carers, the food deliverers, the cleaners, the doctors, family members…”

            If they’re workers, they’ll probably need to stay home and as clear of other family members as possible, if that’s what you mean. We have various ages in our family, so far, so good, wear a mask, keep the contact down and the distance up, but no one is totally isolate. They don’t have to be totally isolated. No one is now.

            I make the suggestion because that’s the at-risk group. With a smaller group and the economy still in tact, more financial support could be available for people who need it, and given that most people in that age group are retired, fewer people would be affected by loss of work.

            In fact you might even be able to provide a full salary replacement and job protection for anyone who’s in the age bracket and has worked in the last year. You could do the same for people with at risk conditions for a lower age bracket, say age 40 and up.

            This doesn’t entirely eliminate the risk for everyone but certainly would take a load off the hospitals and get other people back to work.

            • jim says:

              Actually the more I think about it the more I like it.

              The salary replacement would be up to some cost of living measure in the area. It might even work on a voluntary basis. Basically you’re offering them a buyout. Make it good until contact tracing is to full speed, so possibly up to 4 months or something.

            • Zhou Fang says:

              The point I’m making is that older vulnerable people require a large number of support individuals to survive, they cannot simply be placed into a stasis bubble on their own. Under the current lockdown there is still essential contact happening, but because there’s broader social distancing we also protect those contact points from getting infected. Targetted isolation is unlikely to work because isolated vulnerable individuals will still have to make contact with particular high exposure nodes, and without universal isolation the chances of those nodes having infection is much higher. Despite the isolation you will retain most of the increased risk.

              • Brent Hutto says:

                That’s true for a lot of older, vulnerable people living on their own to one extent or another. And it’s doubly or triply true of those in assisted living or skilled nursing facilities. A huge number of people must come and go, three shifts a day, in those facilities. Each one of them is living outside the “bubble” and shows up for each shift potentially carrying whatever they picked up while off-duty.

                But once again, the problem of “for how long” rears its head. If there are 250 residents in a facility being protected by screening caregivers, caregivers wearing masks and gloves and everything else that make since to keep an infection from sweeping through the facility then that “bubble” will still have the same 250 equally vulnerable people next month, next year, probably the year after that.

                Of course a highly effective vaccine or some degree of herd immunity could miraculously appear a year or so from now. But more likely we are going to have to either find an alternative way to keep vulnerable populations protected or else we will have to accept ongoing large-scale losses within those populations.

                I wish I could find some quantum of encouragement that an effective, universal program of testing/tracing/quarantine might be only a few months away but nothing I’ve seen indicates that anything other than a pipe dream and a political talking point.

              • Zhou Fang says:

                I would say that geographically restricted reopening makes much more sense than sectoral/demographic restrictions. If a town has no cases for a while, and we think we can isolate any infection coming in from outside of the town, then we can probably re-open.

              • jim says:

                Deliveries don’t seem to be an issue. I think that much is clear so far.

                Across the broader population masks, social distancing, working from home wherever possible, and other low cost measures can be maintained.

                Nursing homes are going to be a problem for the foreseeable future. They’ll have to rely on their employees to social distance and if possible provide daily testing capacity.

                See my comment on testing below.

              • jim says:

                Current data in WA state:

                > age 60 = 91% of fatalities.

                That’s 57 total fatalities for people under 60 in a population of 7M.

                That’s just ***SCREAMING*** to be addressed.

    • Stephen M St Onge says:

      The Case Fatality Rate figures are dubious, for reasons that I_ think should be obvious, but hardly anyone I’ve seen notices.

      Hypothetically, 1,000 people are diagnosed with Covid-19 on March 1st. On March 8, 985 are still sick, ten are recovered, and five have died. What’s the CFR?

      Simple. 5/1000 = 0.5%. Serious enough, but not catastrophic.

      No wait, the CFR is 33%, 5/15 died, 10/15 recovered. PANIC!!!

      It seems intuitively obvious that 33% is too high. The people most likely to die in one week are the sickest and most fragile. But it seems even more obvious that the rate rather greater than 0.5%. Do you really believe none of those 985 is going to croak, that all are going to recover?

      We can look at the recovered to deceased ratio over time and see if it changes, but we really need to concentrate on those in our original sample. The case fatality rate should be decreasing with time, as experience improves treatment. But as long as the exponential increase phase continues, the number of new patients swamps the existing cases. If the average time from infection to closure is 15 days, and the case-load doubles every 2-3 days, by the time our 985 are all resolved, 32,000 to 181,000 will be diagnosed, and our original group is essentially noise in the data. And the average case duration may be rather longer than 15 days.

      By now we know that there are asymptomatic and mildly-symptomatic cases. IF the proportion to the diagnosed cases stays stable, say 15 asymptomatic or mild to 1 diagnosed, we get the true infection fatality rate by dividing deaths by 16. But what about people who should have been diagnosed, but weren’t, and then died? They raise the fatality rate. But how much?

      Looking at the recovered-to-died ration over time on Worldometer, and assuming the diagnosed-to-infected ratio is reasonably constant, I think we get an Infection Fatality Ratio of about 2%. This is not apocalyptic, but it is pretty bad, and higher than most estimates.

  12. jim says:

    The article is fine as far as it goes but it’s not providing anything revolutionary that I saw. Most of the numbers are little more than wild guesses. Seems to me like chances are good a lot of things in production will come on line at the same time and a flood of supplies will appear apparently out of the blue. When is the big question.

  13. Tom says:

    FWIW, the linked article for the claim that the virus is disproportionately killing low-income people is that low-income grocery workers are essential and interact with lots of people who need to buy groceries – not that they are dying in droves. It doesn’t actually provide any evidence for the claim. In any case, I think it is true primarily because low-income people probably have more pre-existing conditions but it is nevertheless unsubstantiated.

Leave a Reply