A Tragedy, Not a Statistic

Aaron David

A fourth generation Californian, befuddled.

Related Post Roulette

25 Responses

  1. Oscar Gordon says:

    The testing is key, and we have not been doing nearly enough of it.Report

  2. Doctor Jay says:

    I want to acknowledge that this is hard on us. It’s hard to stay at home. It’s hard to tell others to do it. It’s hard to see your beloved comic book store close its doors with the thought it might not come back. It’s hard to see your favorite restaurant, the one you have been going to twice a week for the past 5 years close. This is just hard.

    And, I have an issue with the framing of “only X people have died”. The problem is that you didn’t include the word “yet”. We’re still in the leading edge of this thing. On our trendline, we will have reached a million deaths by April 26 (https://www.motherjones.com/kevin-drum/2020/03/coronavirus-growth-in-western-countries-march-27-update/).

    Yes, that’s math. I’ve seen remarks, though not from y’all, of “that’s just math, that’s not real”. It makes me shudder. Math put men on the moon.

    On this trendline, we will get to a million deaths in three weeks time. That’s how fast this is growing. At a million deaths there might be maybe 3, maybe 5 million serious cases, requiring a ventilator? So a couple thousand unused will be used up.

    Predicting this course is not as reliable as calculating trajectories of rockets, but it’s pretty solid. The main thing that will knock us off this trajectory is our own interventions.

    We are finishing up our second week of sheltering in place in our county. I’m unsure how long this will continue. I’m quite clear that it is helping a lot. Santa Clara was mentioned as a hot spot three weeks ago. I am willing to do this, because not only does it keep me and my family (my daughter has MS!) safe, it keeps me from being the guy who spreads the virus to 100 other people at a corporate meeting, not knowing I’m even sick. (That happened!)Report

    • Oscar Gordon in reply to Doctor Jay says:

      The problem as I see it is trendlines and a lot of the models predicting those trendlines are (relatively) simple linear models, with lots of baked in worst case assumptions.

      This is why the testing is so damn important, because it helps eliminate those assumptions and improves the modeling.Report

      • Kazzy in reply to Oscar Gordon says:

        I (stupidly) got into a fight with a local guy who posts feverishly on FaceBook as an “expert”. He has two speeds: “The numbers are bad and can only get worse” or “The numbers are not-so-bad which means they must get worse.” He claims to have a stats background (maybe he does) but he never offers any actual numbers or analysis. For whatever reason, he is committed to the worst case scenarios and will just force the data to support them. He’ll say something like, “Assume 50% population infection,” and when I challenge that assumption, he just insists we have to assume something. This is a stats guy?! Sure, certain assumptions need to be made, but there are ways of coming up with those assumptions and the ability to run models multiple times and multiple ways with different assumptions to get a range of outcomes. Not for this guy: bad and getting worse or not-so-bad and getting much worse. Only possible routes.

        I shared elsewhere. My ex has it. Confirmed diagnosis. (She is doing well and likely will be out of the woods entirely in the next two days). My sons were with her during her contagion period. They then came to me and have been with me for two weeks. None of us show any symptoms, so we can’t get tested. My girlfriend and her daughter were with us for much of the two weeks. Same thing: no symptoms and, therefore, no test.

        Right now, this all gets thrown into the data as 1 confirmed case that did not require medical intervention). But it is possibly as many as 6 cases, 5 of them asymptomatic. Or it is 1 case with zero transmission despite intense interaction between parties. Or it is somewhere in between and a mix of both. We just don’t know because the testing isn’t available to us.Report

        • Mike Schilling in reply to Kazzy says:

          Testing is the key. Without that we’re just speculating about all the key numbers: infection rate, transmission rate, mortality rate, etc.Report

        • Doctor Jay in reply to Kazzy says:

          There is a chance that you are an asymptomatic carrier. I hope not, but this is something that happens.

          If we had the testing resources, you would be tested, and if you were a carrier, you would be isolated.

          Having said that, I don’t necessarily disagree with the triage decision to not test you.
          Yeah, we need more testing. A lot more testing.

          And here’s hoping that you and yours stay well!Report

          • Kazzy in reply to Doctor Jay says:

            We were already pretty tightly quarantined (only leaving for outdoor exercise and groceries) and have tightened further (no groceries… leaving property only half as often for exercise) due to possibility of asymptomatic carrying. We are cleared on Tuesday.Report

      • Doctor Jay in reply to Oscar Gordon says:

        Top line, I endorse this wholly. We need to be doing a lot more testing. We need to get to a ‘test and trace’ scenario. We are slowly getting there, but we aren’t there yet. So what do we do until we get there?

        There is a lot we don’t know, I accept that. And there are plausible scenarios where things get very, very bad. They might or might not be correct, but I’m certainly not willing to assume they aren’t the case. That’s terrible risk management.

        At the same time, of course the disease will spread during the initial phases following an exponential growth path. How could it do differently? All the data we have says it is doing so.

        It might not be so bad, but then it might. We might see 3 million deaths, or it might be 300,000. A lot depends, a lot, on what we do. I am willing to do my thing to keep all y’all and myself safe. We’ll sort out the money stuff. We can figure that out, if we’re still here.Report

  3. I think I disagree with the OP, for many of the reasons Dr. Jay said. I will confess, though, that there’s a lot I don’t really know. People who have a better grasp of the necessary skills (statistics, public health measurements, etc.) can speak to how much the OP is right and how much it is wrong. It’s likely not entirely wrong or entirely right.

    This OP implies (maybe not intentionally?) that our current efforts, such as they are, are an overreaction. I certainly hope so!

    One final note: if I read the OP right (and please correct me if I”m wrong), it seems to suggest the number of deaths in Italy ascribed to covid19 is inflated. My question, though, is, is the number of deaths in Italy abnormal? Is this just the comparable number of deaths we’d expect any time in Italy? Are tales of bed-shortages and ventilator shortages in Italy exaggerated or nothing unusual? It seems to me the answers to those questions are yes, no, and no. If I’m right about those answers, then a robust response to the covid19 pandemic seems called for.

    Again, I hope you’re more right and I’m more wrong. But it doesn’t seem that way.Report

    • One more thing to add. While you’re going to get A LOT of pushback on this post (my comment being part of that), I do realize this is a time of uncertainty and we’re trying to figure out basic things about what is going on. It’s a fog of war type of situation. We’re all groping for answers. I don’t fault Aaron for advancing his argument, even though I disagree with him quite strongly.Report

      • greginak in reply to Gabriel Conroy says:

        Prelim info/study suggesting that Covid death toll may be higher then understood. Very plausible but needs more study there and here.Report

      • Dark Matter in reply to Gabriel Conroy says:

        If he’s right then we need good explanations for multiple numbers. The increase in cases is because we’re testing more. The death rate is because we’re putting things into that bucket which shouldn’t be.

        If Aaron is right then we WON’T see a sharp increase in the death rate soon which could have been avoided by vents.

        And Trump will turn the economy back on for Easter.

        If he’s wrong then New York is going to have a nasty few weeks with morgues over flowing, with the rest of the country duplicating that experience aftwards.Report

  4. InMD says:

    I’d be somewhat less concerned if my job didn’t give me a window into the healthcare system and delivery specifically. Our infrastructure in a lot of places is stretched and under a lot of pressure in the best of times. We aren’t exactly going into this with our best foot forward. A shock that’s manageable in theory could wreak a lot of havoc in practice, and that’s not even getting into the way benefits of all kinds are still overly tied to employment.Report

  5. Slade the Leveller says:

    Correct me if I’m wrong, but wasn’t there a study (warning: large PDF) done in Britain to test the do nothing theory? Indeed, there was. The conclusion of the study was this:

    In the (unlikely) absence of any control measures or spontaneous changes in individual behaviour, we would expect a peak in mortality (daily deaths) to occur after approximately 3 months (Figure 1A). In such scenarios, given an estimated R0 of 2.4, we predict 81% of the GB and US populations would be infected over the course of the epidemic. Epidemic timings are approximate given the limitations of surveillance data in both countries: The epidemic is predicted to be broader in the US than in GB and to peak slightly later. This is due to the larger geographic scale of the US, resulting in more distinct localised epidemics across states (Figure 1B) than seen across GB. The higher peak in mortality in GB is due to the smaller size of the country and its older population compared with the US. In total, in an unmitigated epidemic, we would predict approximately 510,000 deaths in GB and 2.2 million in the US, not accounting for the potential negative effects of health systems being overwhelmed on mortality.

    In short, to quote a darling of the right, “facts don’t care about your feelings.”Report

    • Stillwater in reply to Slade the Leveller says:

      not accounting for the potential negative effects of health systems being overwhelmed on mortality.

      That’s the kicker, seems to me. I haven’t seen hard data on this but it seems pretty apparent that death rates are primarily a function of hospital capacity. So they’ve left out a *very* significant factor in their analysis.Report

      • InMD in reply to Stillwater says:

        I’d expect the tussle over resources to get severely nasty. High density economically and culturally important areas would be demanding special attention while rotting hospitals across the deep south and rust belt could end up looking like 3rd world disaster zones. These places were already going bankrupt left and right before this started.Report

      • Slade the Leveller in reply to Stillwater says:

        I think you’re misreading this. I think the authors are saying the large death toll would happen with an optimally functioning health system.Report

        • Stillwater in reply to Slade the Leveller says:

          I don’t think so?

          They’re saying the unmitigated spread of covid would result in massive loss of life *with* a fully functioning healthcare system, and that their model doesn’t include increased mortality projections due to healthcare system failures. IOW, the death rate in practice would be worse than their model predicts.Report

    • Aaron David in reply to Slade the Leveller says:

      He said that expected increases in National Health Service capacity and ongoing restrictions to people’s movements make him “reasonably confident” the health service can cope when the predicted peak of the epidemic arrives in two or three weeks. UK deaths from the disease are now unlikely to exceed 20,000, he said, and could be much lower.


      Niel Fergusson is the first name on the paper you link to, and he has walked back the clames of 500k when questioned in front of parament. To be fair, he has waffled back and forth a bit in the time since testifying, which kinda puts him as a less that reliable source. At least in my book.Report

  6. Urusigh says:

    Thank you for putting forward a reasonable, empirical perspective that is much lacking in the conversation these days.Report