Namecalling in science

A couple of days ago, I retweeted an article with the title [t]he trouble with ‘Covid denialism’. I thought the article was reasonable, but some objected to the use of ‘denialism’. There are a number of very credible scientists who have promoted some of the scientific ideas being highlighted in this article and we should, ideally, treat these alternative scientific ideas with some respect and should engage in constructive discussions, rather than suggesting such people are in denial.

However, having been engaged in the public climate debate for quite some time, I’m somewhat inured to the use of denial, or denialism. On the other hand, this may be an opportunity to discourage its use at an early stage of a topic so that we can maybe encourage more constructive dialogue. I would certainly be in favour.

Unfortunately, my experiences in the public climate debate has also made me rather cynical. Certainly, in the climate context, the use of denial, or denialism often seems entirely justified and the complaints are typically more related to people trying to deligitimise their critics, than a genuine desire to improve the dialogue. It can be a way to try and get the freedom to promote contentious scientific views in public without being criticised. Similarly, in my experience in the climate debate, those who complain most about tone are often people who are quite comfortable being pretty blunt when it suits them.

I also tend to think that scientists shouldn’t expect the same level of engagement in the public sphere as they might in a more formal scientific setting. Scientists shouldn’t expect some kind of special treatment in the public sphere just because they’re scientists. They should, of course, get the same level of legal protection as anyone else who is engaging publicly, but they shouldn’t expect to not be judged for the views they choose to express, or the groups with which they choose to associate.

Of course, our understanding of Covid-19 is not nearly as mature as our understanding of anthropogenically-driven climate change. Hence, maybe we should avoid throwing around labels at this stage and maybe we can still engage in constructive dialogue with those who promote scientific ideas with which we disagree. You’ll excuse me, though, if I’m not confident that we’ll actually do either.

Links:

Guest post : Label the behaviour, not the person – Guest post by Richard Betts.

This entry was posted in ClimateBall, Personal, Philosophy for Bloggers, The scientific method, We Are Science and tagged , , , . Bookmark the permalink.

170 Responses to Namecalling in science

  1. brigittenerlich says:

    It is interesting to see the emergence of labels like covid sceptic, covid contrarian and covid denier which parallel similar labels in the climate change debate, although in the covid debate the denier label seems to have the upper hand at the moment. There has been a lot of discussion in climate communication circles around labelling and it would be interesting to do similar research in the context of the current pandemic and covid communication. From what I can see, there are interesting differences underneath the activity of labelling in the two domains…. but more research needed, indeed a lot of research needed…..
    One thing to keep in mind regarding covid ‘scepticism’ is perhaps that one might want to distinguish between ‘organised covid scepticism’ which is the life blood of science; ‘active covid scepticism’, which at the heart of (science) communication and deliberation, and ‘special interest covid scepticism’ which has affinities with and perhaps feeds on and into denialism.

  2. Jim Hunt says:

    “Label the behaviour” – Richard Betts

    Isn’t that what Unherd is doing?

    “There are, however, some people on the lockdown sceptic side of this debate who would deny the existence of this tradeoff; they are flirting with a sort of ‘Covid denialism’. The pandemic, they claim, if it ever existed, is over because ‘herd immunity’ has already been reached.”

  3. Brigitte,

    It is interesting to see the emergence of labels like covid sceptic, covid contrarian and covid denier which parallel similar labels in the climate change debate,

    Yes, I found this interesting to.

    One thing to keep in mind regarding covid ‘scepticism’ is perhaps that one might want to distinguish between ‘organised covid scepticism’ which is the life blood of science; ‘active covid scepticism’, which at the heart of (science) communication and deliberation, and ‘special interest covid scepticism’ which has affinities with and perhaps feeds on and into denialism.

    Yes, I agree. Context matters. Labelling people pejoratively when they publish a paper with a result one doesn’t like would be very much against the scientific norms. Using a label to describe someone’s association with an organisation that has a clear agenda might, for example, be perfectly reasonable.

  4. Jim,
    Yes, I think they were mostly doing as Richard suggested in that earlier post. It was more labelling the behaviour than labelling the individual – “flirting with denialism” as opposed to “they are a denier”.

  5. Brigitte,

    There has been a lot of discussion in climate communication circles around labelling and it would be interesting to do similar research in the context of the current pandemic and covid communication. From what I can see, there are interesting differences underneath the activity of labelling in the two domains…. but more research needed, indeed a lot of research needed…..

    Just out of interest, what process would one follow in doing such research?

  6. brigittenerlich says:

    That is a good question. I am starting my usual media analysis but will have to interrupt that soon for a while – so here one would use qualitative methods like thematic or discourse analysis or metaphor analysis and/or more quantitative computer assisted ones like corpus linguistics. One could, of course, also look at Twitter and so on using various Twitter analysis methods, semantic maps, network analysis etc.. If one had time and money one could probably also do surveys and focus groups etc etc. and analyse them similarly. I have just, about half an hour ago, discovered a freshly published book on ‘Climate change scepticism’ in Germany, France, the UK and the US – from an ‘ecocritical’ perspective. Just ordered it to see what that means and how that translates into any method – and whether I can learn anything regarding the study of covid scepticism. https://www.bloomsbury.com/uk/climate-change-scepticism-9781350057043/

  7. Brigitte,
    I wrote a post about that book. I also exchanged a number of very pleasant emails with the lead author (Greg Garrard). His brother is an astronomer in the UK who I have met, but don’t know well, but who has been helping one of my PhD students with his project.

  8. brigittenerlich says:

    Ha, how did I miss that!!!!

  9. You mean you don’t thoroughly read every single one of my posts? 😀

  10. mark4asp says:

    You cannot actually define COVID denialism, or COVID-19 denialism, but you may use it because it’s a badge of identity for you – opposing “denialists”. The article you tweeted was actually about “lockdown denialism” written by an economist and a psychologist. It’s not titled ‘Lockdown Denialism’. But it’s actually about that.

    “‘lockdown sceptics’ have rightly argued, these measures cause immense economic, educational, social and health damage, and undermine our civil liberties. We are, for now, stuck between a rock and a hard place.”

    “some people on the lockdown sceptic side of this debate who would deny the existence of this tradeoff; they are flirting with a sort of ‘Covid denialism’.”

    Initially the authors don’t actually name a single person who denies a cost-benefit trade-off. This is a classic attempt at formulating a “moral panic”. A well-know technique in politics for over 50 years now. After they identified the social pathology: “COVID denialism” they fit someone up, like incompetent cops solving a murder by pinning the blame on some minority or marginalized person. In this case: “well-credentialed Dr Michael Yeadon, a PhD in respiratory pharmacology” who very conveniently is also “a former leader in pharmaceutical companies”. Excellent – recruit the pharm-hatred, and anti-capitalists too to one’s moral panic! Very astute, from a writer at a well know neo-Liberal think-tank! Maybe Yeadon is guilty as charged – guilty of errors of logic. But is guilty of being a spokeman for “COVID denialism”. No, because that’s not actually a thing,

    Scientifically, it’s an odd approach. First claim a moral malaise in society: “COVID denialism”. Then back your claim on in a single article by a single person. It’s the opposite approach to how science is done. In science we can legitimately generalize when we have some data to generalize with: when we’ve identified many writers, more or less saying the same thing, we can say there’s a group of them, and label them as “COVID denialists”. Not what’s done here. The essay you cited is political demonization, an attempt to formulate a moral panic.

    Ref.: Stanley Cohen, (1972) ‘Folk Devils and Moral Panics’

  11. mark4asp says:

    Ooops my bad English. No edit button

    “recruit pharm-haters, and anti-capitalists too to one’s moral panic! Very astute, from a writer at a well know neo-Liberal think-tank! Maybe Yeadon is guilty as charged – guilty of errors of logic. But is he guilty of being a spokeman for “COVID denialism”. No, because that’s not actually a thing – it’s a political slogan.”

    However – I can see why you like it – preferring political slogans in ‘raising awereness’ to scientific arguments, as you do.

  12. Mark,
    I wasn’t specifically wanting to defend the article, or promote the use of terms like “covid denialism”. I was mostly highlighting the similarity between what has happened in the climate debate, whether or not we want to discourage this in the covid context, and my cynicism about whether or not this would actually work, and whether or not it is really actively worth pursuing. A point I’ve made before, but didn’t make here, is that the only person’s tone you can really control is your own. If someone thinks that the debate should be constructive and polite, then – IMO – they should start by setting an example of how they think we should engage in discussions.

  13. Jim Hunt says:

    Mark,

    You seem to be making a lot of (invalid?) assumptions in your (overly?) hasty responses?

  14. Mark’s comment about I can see why you like it – preferring political slogans in ‘raising awereness’ to scientific arguments, as you do
    reminded me that I should have included Joshua’s comment about the issue of namecalling 😉

    If only those poopyheads would stop calling us poopyheads.

  15. Well, it is obvious we learned nothing from a decade of the climate conversation. Sorry, Richard Betts–maybe next time.

  16. Jim Eager says:

    Oh, punkin, saying there is not a single piece of empirical evidence and no supporting known physics does not make it so. What it does make is a prefect example of denialism.

  17. Tom,
    What would you take away from a decade of the climate conversation?

  18. Joshua says:

    Anders –

    Was going to make that observation but I see you’ve included it in the comments,

    Mark –

    From the article…

    > “‘lockdown sceptics’ have rightly argued, these measures cause immense economic, educational, social and health damage, and undermine our civil liberties. We are, for now, stuck between a rock and a hard place.”

    This bugs me because it’s a classic example of confusion between correlation and causation.

  19. ATTP, that a very large percentage of participants on both sides were more interested in deligitimizing the other side than discussing what science was in the process of revealing.

    As it is perfectly legitimate to hold opposing views on the consequences of what science was (and is still) revealing, the conversation turned into mud wrestling on one side, Climate Ball on the other.

  20. Joshua,
    It seems clear that lockdowns do have economic, educational, social and health implications. I’m not so sure that I think there’s civil liberties issues, but I do agree that lockdowns can do a lot of harm. However, not doing something that might reasonably be described as a lockdown can also do a lot of harm, and may also have economic, educational, social and healthcare implications. So, what correlation/causation claim are you taking issue with?

  21. Joshua says:

    > and undermine our civil liberties.

    That also bugs me. Do the citizens of New Zealand or South Korea or Taiwan feel that their civil liberties have been undermined, or that they’re less likely to get sick or die?

  22. To apply the lessons learned to discussion of the social science surrounding Covid-19, it would be well to state very clearly the factual basis that should underpin the discussion:

    We have clear examples of how societies succeeded in controlling the spread of the virus.

    1. Isolation of an entire population, as achieved by Taiwan and New Zealand
    2. Test and trace of as many people as possible, with quarantine of positives and their contacts

    Everything else is a discussion of how best to fail gracefully. As failing gracefully is now the unstated policy of almost all countries it should be clearly stated–protect yourself and your family as best you can until a credible vaccine is available.

    That should be the starting point.

  23. Tom,
    I broadly agree, but (I think) I spent a long time (unsuccessfully) trying to discuss the science. Even though there are things I would have done differently had I known then what I know now, I don’t think the outcome would have been much different. So, I don’t really think that if we avoid labelling people in this context that it will lead to a more constructive discussion. I’m not suggesting we not try, I’m just not convinced that it will make much difference.

  24. Joshua,
    As far as I can tell, New Zealanders are quite pleased that they seem to have dealt with this quite effectively.

  25. Joshua says:

    Anders –

    >… social and healthcare implications. So, what correlation/causation claim are you taking issue with?

    First, I don’t see any reasonable way to disaggregated the impact of the virus itself from the interventions enacted to address them.

    As an example – not conclusive but if interest:

    https://ourworldindata.org/covid-health-economy

    Second, the logic rests on assumptions about counterfactuals. What would the economic and social harm have been absent the interventions? For all we know, they could have been worse.

    Third, treating harms as if they don’t manifest a differentially across factors such as SES, baseline health, etc, is of limited value. For example, SIPs and NPIs enabled people to get stimulus checks and collect unemployment whereas absent those policies they could have been fired without those advantages if they decided to stay home from work because they had a comorbidity.

    That isn’t to say that we should ignore the potential of harm, say, from closing schools – especially with an eye towards the differential effects of the SIPs and NPIs in the other direction.

    But I think that in general, people are too quick to assume here that correlation equals causation.

  26. Willard says:

    Well, it is obvious that our usual concern merchants have learned nothing from a decade of ClimateBall. Sorry, AT–maybe next time.

  27. Joshua says:

    I’m concerned that the name-calling of alarmists is causing a problem.

  28. Joshua says:

    For me, the most salient parallel is how it illustrates how difficult people find it deal with risk in the face of uncertainty, especially with low probability high damage function risk.

    Suppose hospitals had become overwhelmed? What might the level of damage have been had that happened? Maybe they weren’t likely to get overwhelmed. But what if they had? The virus spreading uncontrolled and basically only a tiny few able to get treatment – with those sacrificing the most (healthcare workers and people who had to be exposed because they didn’t have the resources to isolate) suffering the greatest harm.

    Dealing with such risk comprehensively requires that we articulate values and negotiate them. The difficulty we have to engage in stakeholder dialog to create policy in the face of uncertain risk is, to me, the most salient parallel.

  29. We should start that dialogue pretty quickly. Hospitals are being overwhelmed as we speak and as a result mortality will quite likely double or even triple starting in about 10 days.

  30. This is analogous to one part of the climate conversation–where many argue that it isn’t as much the total scope of climate change that is most dangerous, but rather the speed with which it is occurring.

    We have developed (in less than a year!) much better treatment protocols for Covid-19. But the rate of new infections during this present spike prevents healthcare professionals from following these protocols with overflow patients, leading to triage and palliative care for many.

  31. Tom,

    We should start that dialogue pretty quickly. Hospitals are being overwhelmed as we speak and as a result mortality will quite likely double or even triple starting in about 10 days.

    Yes, I agree. However, my experience in the climate context is that it’s much easier to say what we should be discussing than it is to get people to actually do so.

  32. Well, I can start with a proposal–every medical professional currently serving in the armed services should be deployed today to a severely impacted hospital.

  33. Joshua,

    First, I don’t see any reasonable way to disaggregated the impact of the virus itself from the interventions enacted to address them.

    Okay, yes, I agree. I think some of the narrative has been unfortunate in the sense that it has pitted trying to control the pandemic against protecting the economy when it seems likely that the two are related. There are economic implications to failing to implement strategies to deal with the epidemic and there are implications to implementing these strategies. It seems likely that the optimal strategy is one that attempts to control the spread of the virus so that we can return to some semblance of normality.

  34. The bottleneck right now isn’t PPE or medication–it’s staffing.

  35. I’m not sure how much of this is happening in the UK, but I have seen pictures of people in military uniforms staffing some of the medical centres.

  36. Joshua says:

    Tom –

    > Well, I can start with a proposal–every medical professional currently serving in the armed services should be deployed today to a severely impacted hospital.

    Seems to me like that’s a good idea. Only problem is it would necessarily require that our federal government acknowledge that there’s a problem and our federal government is laser-focused on not acknowledging that there’s a problem. And about 1/2 of our fellow citizens don’t want to acknowledge that there’s a problem because that would imply that our federal government has in some way failed to address a real problem.

    What a mess.

  37. “Of course, our understanding of Covid-19 is not nearly as mature as our understanding of anthropogenically-driven climate change. “

    Smart to put the modifier on there, since an understanding of natural climate change over the span of more than one year is still highly immature. Medical science was able to isolate the virus rapidly — essentially as it started to spread — whereas even after decades of study, climatologists still have not isolated the root cause of climate dipoles and thus unable to make any predictions of yearly swings in temperature extremes.

  38. Paul,
    Can we make that your one drive-by per post?

  39. mark4asp says:

    [Enough #ButCAGW, Mark. -W]

  40. Joshua says:

    Hey Steven –

    I rember a couple of years ago we had a,… uh… discussion about Nic being referred to as an independent climate scientist.

    Granted, he was at that longer but weould you now call him an independent epidemiologist or independent virologist? If not, what’s the dividing line?

  41. Rob Kenney says:

    My takeaway from the linked article was that a reality check should guide policy.
    The authors showed Yeardon’s scientifically-based claims to be unsound, although the metrics of harms related to lockdowns initially gave rise to the “denier” label.
    In the mainstream what occurs is that some latch onto Yeardon’s views to legitimise their actions/beliefs as they are accepting of arguments from authority. Yeardon need not be labelled anything other than a science professional. Indeed, that label supports those in the mainstream who choose his words to now actively deny what in reality is thin or baseless, or even plain wrong.
    It’s deja vu climate change denial 101 as I see it.

  42. Joshua says:

    -snip-

    The harsher measures taken by the other Nordic governments (all but Norway’s headed by Social Democrats) have resulted in soaring support for the governing parties there — more so than in Sweden. In Denmark, Norway, and Finland, support has been hovering between 65 and 88 percent since April, the highest of all governments in Europe, while Sweden’s never surpassed 65 percent. Moreover, the public in other Nordic countries has been highly exposed to reports on the Swedish strategy, sometimes presented as a competition between the countries. Regular comparisons show death rates, infection numbers, and admissions to emergency wards in the different Scandinavian countries — and this information is also displayed side by side during live TV debates on the pros and cons of the diverging approaches.

    -snip-

    https://jacobinmag.com/2020/11/sweden-coronavirus-covid-nordic-scandinavia

  43. Steven Mosher says:

    “Granted, he was at that longer but weould you now call him an independent epidemiologist or independent virologist? If not, what’s the dividing line?”

    No I would still call him an independent scientist. what he does is data analysis.
    Today it’s socially accepted practice to call these people data scientists. or scientist
    for short. Meh.

    What he lacks is subject matter expertise. So if hewas a SME in climate and did data analysis
    I would call him a climate data analyst or climate data scientist.
    He has no Subject matter expertise in Epi or virology. witness how he (and willis and others)
    used the Diamond princess without even knowing to check GISAID for which genetic isolate
    those folks were infected with ( the less infectious strain).
    He built a model and ran it. So, data modeler, or data scientist ( like my friend Marc Bevand)
    When Levitt ( nobel prize winner) does bad epi does that make him a non scientist?

    Your difficulty I think is ascribing too much value to the socially constructed label.
    I call Levitt a scientist, cause that’s how society has chosen to label what he does.
    I’d call Marc or Nic data analysts or data scientists, or scientist. cause that describes
    for me what they do. They are not ditch diggers.

    Because the label is pretty broad and because I don’t give brownie points for it, that
    means I am forced to look at the actual work. I look at Marc’s code, look at Levit’s
    math, look at Nic’s code/math. yup, data analyst, data science: science; not ditch digging.

    Now I think you want to use the label more restrictively like a badge. and that badge helps you
    decide who to believe. me? I don’t care about the badge. I care about the work. You
    are welcome to use the word differently than I do. That’s how language works.

    Dividing line? as in a binary function?. Let see. I’m not too keen on
    dividing lines in language ( it’s a binary thing)
    Scientist, for me, is at the bottom just a label people attach to a varied class of behaviors
    and professions over a changing universe of selected subject matter that DOESNT NECESSARILY have a canonical criteria for inclusion/exclusion.
    Think of the differences between cups and mugs and glasses and brownie points
    if you get the reference to the work in linguistics I am referring to.

    So we end up with data analysis becoming data science and computing becoming
    computer science, and politics becoming political science. Do they all share some
    feature, some ‘real’ nature at the core? probably not, language don’t work like that. If you
    demanded that I propose a “defining characteristic” of sciencing behavior it would be vaguely
    “explaining stuff about the physical world in quantitative ways”
    To be an independent epidemiologist, would probably require some SME in epi.

    One problem with doing data science is that you think that you can just jump in and
    “understand” the data. That’s the mistake that Nic and Nobel prize winning Levitt ( and others ) make. Those mistakes don’t make them non scientists. It makes them mistaken.
    I’t also one reason I don’t do any charts on this stuff.

  44. Everett F Sargent says:

    I am too busy listening to David Gilmore …

    … to partake in this deeply intellectual discussion. 😉

  45. Steven Mosher says:

    “Well, I can start with a proposal–every medical professional currently serving in the armed services should be deployed today to a severely impacted hospital.

    In Singapore military staff was deployed to do contact tracing.

  46. Steven Mosher says:

    “Second, the logic rests on assumptions about counterfactuals. What would the economic and social harm have been absent the interventions? For all we know, they could have been worse.”

    It astounds me the number of people, really smart people, who don’t get or selectively don’t see the issues of counterfactuals.

    it’s like they took crazy pills

    Anyway, here in Korea we have 57 people in critical care. Shockingly there only 130 open beds
    in ICU for a nation of 50M. Let that sink in. We continue to have clusters in nursing homes
    and hospitals. So they will go in and do 100% testing at all nursing homes and remove the sick.
    masks are now mandatory and there will be a fine for not wearing.

    They are moving to add 300 more beds by next year??? Interesting tidbit from China. During the
    last SARS China build a hospital in Beijing dedicated to SARS. when that ended they mothballed
    the hospital and early this year they re opened it for COVID.
    The market never would have built that hospital

    last tidbit: the typical flu deaths for Korea is around 20K. Covid is at 500.
    Nobody here does any sketchy math around excess deaths.
    more people died while taking the flu vaccine this past month( ~100) than died from covid.

  47. Everett F Sargent says:

    I thought that this was an open thread, where we all could do some namecalling for the sake of science or maybe the STS of namecalling in science. But namecalling, in general, is everywhere and has always existed last time anybody checked. Insert (Monty) Python script here >.<

    I mean, a civil discussion only occurs when someone doesn't want to knock someone else's block off. And we already have proof of that when discussions take place when hiding behind screen names on these interboobnets.

    I mean seriously, just tell them what you really think of them, already. Face to face and in person even.

    I am thinking, that if everyone really knew everyone else, that the murder rates would go way up, like really fast even But, you could call me crazy for expressing myself that way or thinking such thoughts.

  48. I don’t have a problem with terms like “COVID-19 denialism”, “COVID-19 denier”, etc. Even though the science on COVID-19 is relatively young, there are various points on it that are so strongly supported by evidence, that refusing to accept them is denialism. I find that many of the people complaining about the use of terms like “denialist” are often the people most willfully ignorant about the relevant evidence, and thus they’re the least likely to know what actually qualifies as denialism on said topic. They’re usually too busy whining about tone to actually address the substance behind the claim that someone is a denialist.

    Whining about someone’s tone doesn’t make what they say wrong; facts don’t care about your feelings, anymore than facts cared about the feelings of those who complained about AIDS denialism (or evolution denialism, or vaccine denialism, or…) being called out for what it is. If people who know more about a scientific field than you point out denialism with respect to that field and that bothers you, then the problem is more likely with you than with them.

    Anyway, below is one of my favorite examples of denialism from Suneptra Gupta, one of the authors of the nonsensical Great Barrington Declaration (courtesy of UnHerd, who are usually, but not always, promoting COVID-19 denialism):

    “Denialist
    A person who does not acknowledge the truth of a concept or proposition that is supported by the majority of scientific or historical evidence; a denier.”

    https://www.lexico.com/en/definition/denialist

    “Yet it would be wrong to prevent the denialists having a voice. Instead, we argue, it is necessary to shift the debate from the subject under consideration, instead exposing to public scrutiny the tactics they employ and identifying them publicly for what they are.
    https://academic.oup.com/eurpub/article/19/1/2/463780

    “The “tone troll” is a real critter. They are the most common subspecies of Internet troll, mostly harmless but super annoying, and easy to spot in the wild: they complain about the tone of a message, rather than its substance.”
    http://www.paulingraham.com/tone-trolls.html

  49. Steven Mosher says:

    “First, I don’t see any reasonable way to disaggregated the impact of the virus itself from the interventions enacted to address them”

    that’s for sure but it is worse than that.
    Even if you could build up ledgers of impacts from deaths, impacts from job loss, impacts
    from people being afraid to go to the hospital, impacts from over run hospitals, damage
    to children from losing out on school, suicide, drug use.. all the direct and indirect damage
    all the collateral damage, even if you could ‘estimate’ all of it, the real questions remain

    A) how do you balance
    B) who decides the balance
    C) how do you enforce the decisions

  50. Steven Mosher says:

  51. Everett F Sargent says:

    “It’s also one reason I don’t do any charts on this stuff.”

    Well. ever since Hansen wrote that God awful paper on ice melt doubling times and before that the exponential growth of sea level rise assumptions. You will come to a much better understanding of the nature of doubling times and misuses thereof.

    So, for example, COVID-19 where we don’t have an effin’ clue as to the missing population dynamics at the very start of the pandemic. By definition, how could we, unless we were to do so on purpose (start the pandemic with various IC’s abd BC’s). Define t = 0 and/or n = 0. so that you are left with taking, not one but, several SWAG’s. That doubling time is very unconstrained in the nearfield but very stable in the farfield (Include/exclude China, for example).

    Fly thousands and thousands of infected from Europe into NYC as Small Hands did? Oh crap, there I go with the namecalling in science again.

    But, in general, take CO2 or SLR and define t = 0 and y = 0 unambiguously and objectively. You can’t, but people do so all the time. Many bad papers get published based on somewhat primitive assumptions that are usually incorrect and grossly oversimplified (calculation of return periods being a classic misuse). If it is semi-log, well how do you define t = 0 (where you start your accumulation or sums), and log-log, that one oh boy you are just asking for trouble.

    So, take CO2, anthro-CO2 atmospheric time series (note that this is already in summation format) in semi-log and do doubling time and you will find that it is currently super-exponential, the doubling time is getting smaller wrt time (and no you can define any numbers to time +/- whatever does not change its semi-log nature of doubling time).

    I’ve only shown the doubling time plots, but in doing so, they then make you think of their much deeper realities and limitations thereof.

  52. Steven,
    At the end of that video Jay Bhattacharya claimed that excess deaths in Sweden were far lower than Covid-related deaths. Have you seen any confirmation of that?

  53. Jim Hunt says:

    Anders,

    There were certainly several Royal Navy personnel helping look after me during my recent stay in Derriford Hospital in Plymouth

  54. Marco says:

    “Have you seen any confirmation of that?”

    I surely have not seen that:
    https://www.nature.com/articles/s41591-020-1112-0/tables/2
    Excess deaths in Sweden is clearly more than reported COVID deaths, by about 20%.

  55. Marco,
    Thanks. Bhattacharya seemed quite confident that the excess deaths in Sweden were about half that of those associated with Covid-19 and seemed to imply that this suggested that the Swedish strategy had worked (I guess because those who died were – according to him – primarily amongst those who would have probably have died anyway). Your article suggests that this is simply wrong.

  56. Joshua says:

    Keep in mind, extrapolating from Sweden to the US or the UK is problematic.

  57. Joshua,
    Of course. I was just checking on Bhattacharya’s claim about excess deaths in Sweden, which appears to be wrong.

  58. Joshua says:

    Steven

    Thx for the answer in Nic.

    > It astounds me the number of people, really smart people, who don’t get or selectively don’t see the issues of counterfactuals.

    >> it’s like they took crazy pills

    I know that conditional probability is hard, but it seems that it is a major vulnerability point for resisting motivated reasoning.

    The counterfactual framing that “lockdowns = death” is treated as axiomatic in every pro-herd immunity argument I have ever seen. Nic wins even deem the issue worthy of discussion. I have only gotten past correlation = causation with one person, over at Gelman’s blog, and his argument boils down to (paraphrasing) “If we had a totally different media that didn’t sensationalize mass casualties from a raging infectious disease, we could pretty much go on as if life were normal during a pandemic.”

  59. Joshua says:

    Anders –

    Yes. But the problem is that Sweden is a rope a dope.

  60. Joshua says:

    Won’t even deem…

  61. Marco says:

    ATTP, there is more information that suggests quite the opposite from what he claims:

    Source: https://www.medrxiv.org/content/10.1101/2020.09.10.20191965v1.full.pdf
    (this is Stockholm County only)
    and
    Source: https://www.folkhalsomyndigheten.se/contentassets/53c0dc391be54f5d959ead9131edb771/infection-fatality-rate-covid-19-stockholm-technical-report.pdf
    (see e.g. Figure 4 on page 19, also for Stockholm)
    Both indicate a higher excess mortality than registered COVID deaths.

  62. Marco says:

    Whoops! Sorry, it automatically embeds the pdfs?

  63. Ben McMillan says:

    These claims that many/most of the deaths attributed to covid are not really due to covid seems pretty outside the bounds of reasonable informed debate at this point. Getting very much into conspiracy-theory style ‘medical establishment manipulating the data’ territory.

    What term would be better than ‘denial’ for the desperate contortions necessary to maintain a belief that this isn’t a deadly disease?

  64. Steven Mosher says:

    Steven,
    At the end of that video Jay Bhattacharya claimed that excess deaths in Sweden were far lower than Covid-related deaths. Have you seen any confirmation of that?”

    No and I didn’t look. My very very limited experience with excess death data ( we did work on air pollution deaths, so I looked at it) In the USA the estimate is seriously lagged. the data is ready for analysis 12 months AFTER the current month. So when Nobel lauret Levitt tried to use it

    later he would admit his mistake

    So.
    1. I would be very skeptical of any excess death data unless it is official and no subject to change and adjustment
    2. It misses several key things

    A) if your constraint is hospital beds you don’t care if excess deaths is lower.
    For example here in Korea we have 130 open ICU beds . at one point we dipped to 80.
    Thats right 80 beds in a country of 50M+
    so if 5000 cases results in 1000 ward beds being filled and 100 ICU beds you are fucked.
    Doesn’t matter if excess deaths is Negative!

    B) It forgets that the future deaths from covid are preventable deaths.

    The “excess death” argument comes in a some different variants, but in the end
    it looks like variant of the “trolley problem”. basically you have a fork in the road and
    folks are asking you to decide which way to go by summing up the harms.

    I’m not the first to say it’s like the trolley problem

    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7224645/
    https://hekint.org/2020/10/15/revisiting-the-trolley-problem-in-the-covid-19-pandemic/

    Point being we are asking politicians to solve a problem that moral philosophers have not. Is anyone surprised that folks are split on this? is anyone shocked that science or modeling
    wont tell you the pain and suffering that lies ahead on either track?

  65. Joshua says:

    Of course, there’s also this (caveats such as “what will it look like in a month or two” and “how reliable are those data?” apply):

    Our World in Data.

  66. Joshua says:

    Hmmm. Any way to fix that link?

  67. Willard says:

    Fixed. WP’s parser isn’t perfect. Sometimes only HTML works.

    I also added “source:” in front of your links, Marco. WP nowadays embed the PDF in pages, like it does with an image.

    Welcome back!

  68. Dave_Geologist says:

    Suneptra Gupta of the Great Barrington Declaration said ~0.05% of people infected w/ SARS-CoV-2 die of COVID-19

    So England’s ~44,000 reported COVID-19 deaths would imply ~88 million infected people

    England only has ~56 million people

    That was Gupta’s problem even back at the time she first published, about six months ago. As I believe I mentioned at the time. While you could fit her IFR and undetected infections to England’s deaths at the time, there were already other countries and hot-spot regions like New York and Lombardy where applying deaths to her IFR meant that more than 100% of the population had already been infected.

    Which meant that either her IFR was wildly wrong, or there is no immunity and you can get re-infected within a month. As opposed to the 6-12 months for common-cold coronaviruses. So if she was right, she’d proved herd immunity by infection, at least the mild infection most get, was impossible. Yet six months later she’s arguing for exactly that. Doesn’t get more denialist: she’s even denying the clear implications of her own study!

    And since then there have been dozens of seroprevalence studies which find around 10% infection and an implied IFR of around 0.6%. I can’t remember how she denies that. T-cells? Or sky-dragons?

  69. Joshua says:

    Suneptra said that the virus in the UK was “on the way out” in May.

    Also, this from early spermember.

    -snip-

    Nothing she has seen since, not even a seeming resurgence of cases, has caused her to reevaluate the scenario. “I don’t think we are seeing a resurgence,” she says. “It’s a useful piece of information to know that lockdown worked in certain areas, to slow down the virus and to stop it from spreading from, say, a town to a village. But as soon as you lift lockdown, the virus comes back.”

    So the natural spread of the virus has merely been delayed? “It’s not really a resurgence. It’s just where it didn’t increase in the first place. Now all the barriers have been removed, it is increasing. I don’t see any surprises in that pattern. What I do think is interesting is that it’s not resurgent in many areas that did suffer the full brunt of the pandemic, so in London, New York, northern Italy, Sweden.” For Gupta, this implies that in these areas levels of herd immunity may have been reached, meaning the spread of the virus is now being contained. The key step now, she says, is to use serological testing to determine what proportion of the population has been exposed to COVID-19.

    Another factor that interests Gupta is the extent of cross-immunity from exposure to other coronaviruses. “I thought it might protect only against disease, but we’ve learned that these cross-immunity responses can actually protect against infection. That makes it hard to use antibody tests to get a measure of how many people have been exposed to the virus. But the good news is that if a fraction of the population is already resistant to infection, that brings the threshold for herd immunity down substantially.”

    -snip-

    So herd immunity was reached in early September in Sweden and Northern Italy. Sure would love to hear her describe how she responds to Worldometers today.

    Entirely sure that T-cell reactivity means immunity from infection.

    Nic thus far had been silent on these questions for quite a stretch.

  70. Joshua says:

    Er, that’s a rough type. Spermember (in case it wasn’t obvious) = September.

  71. Dave_Geologist says:

    At the end of that video Jay Bhattacharya claimed that excess deaths in Sweden were far lower than Covid-related deaths.

    Marginally: COVID-19 and Excess All-Cause Mortality in the US and 18 Comparison Countries.

    In the rounding I should say. 50.8 per 100k vs. 57.4. And disingenuous to my mind. Like the UK MPs saying “but there have only been six deaths in Cornwall in this wave”. Early on Belgium were also reporting more Covid deaths than excess deaths (110%): A pandemic primer on excess mortality statistics and their comparability across countries. Now it’s flipped. And they had a hard lockdown. And I bet they’re not patting themselves on the back over the few thousand excess deaths they “avoided” by not being harder. I recall talk at the time that they had a broader definition of a Covid death than most. More telling is Norway, which locked down hard and fast and shares a 1000-mile border. 5 Covid deaths per 100k, -2.6 excess deaths (that’s minus 2.6) per 100k. If Sweden had only 25k excess deaths he might have a case. But even then the bottom line is that Norway made a much, much better bargain than Sweden. Would you sacrifice 50,800 people to save 6,600? A bad bargain even if you posit Norway’s number is really zero excess deaths because that gain is small enough to be cancelled out by extra cancer etc. deaths in the pipeline.

    At a finer level of detail IIRC Scotland has a slightly broader definition than England. From an epidemiological POV you surely want the broadest possible, so that every possible Covid death is identified for future study. For example early on if we’d been strict we’d have excluded strokes, embolisms and heart attacks in people with a positive test because it was a few months before we knew about the blood-clotting side-effect. Now I hope someone is trawling back through those early cardiovascular deaths for clues to long Covid impacts. It’s easy to say: car crash, person had a positive test but not Covid, than: previously unidentified medical condition, person had a positive test but not Covid. At least in the UK, and IIRC Australia, we basically had no flu season this year because Covid lockdown stopped the usual spread, especially in the first wave when schools were closed. Sweden distanced too so probably got some of those benefits.

    And there will have been some deaths which would have happened anyway, some of which have Covid on the death certificate, but don’t appear as excess deaths because they’re part of the old-people-dying-in-winter factor in cold-temperate countries.

  72. Dave,
    I’m not sure where the 50.8 comes from?

  73. Sorry, I missed the second table.

  74. Steven Mosher says:

    “So herd immunity was reached in early September in Sweden and Northern Italy. Sure would love to hear her describe how she responds to Worldometers today.”

    they just shift the debate.

    First step: lockdowns don’t work ( ignore the counterfactual) or are not needed (Sweden)
    Second step: they may slow the spread, but the collateral damage is worse.

    She is now collecting a database of worldwide damages from lockdowns.

    It would be more interesting if there were a covid lukewarmer wing

    1. Lockdowns work, masks work, testing works, death is real..
    2. Lockdowns also cause collateral damage.

    we need king Solomon

    I think the vast majority of “pro lockdown” folks recognize that there is incalculable damage
    done by lockdowns. Medical care not sought, economic damage, increase drug use,
    suicide, divorce, damage to kids not in school ( haha so much for home schooling being great)

    so maybe someday folks will have the discussion about weighing the damages of both paths
    probably not. it’s an ugly discussion

    you could always try wuhan style for 14 days.

  75. Steven Mosher says:

    sometimes state propaganda is a good thing

    Ask yourself why people never encouraged NYC the way China encouraged Wuhan
    Did you smile when the idiots in North and south dakota started to get sick?
    use it as a tool to make political points?
    Did you blast Cuomo for not being perfect?

    Ya, sometimes state propaganda is a good thing
    “love and hope will always spread faster” ( evil CCP right there)

    ( music is stolen from a korean movie called “the classic”)

  76. Steven Mosher says:

    Joshua

    https://www.dn.se/sthlm/var-femte-stockholmare-som-testas-ar-covidsmittad/

    sweden positivity is up to 20%

    so much for the “low prevalence” false positive meme

    none of this will change minds, however

  77. Everett F Sargent says:

    “none of this will change minds, however”

    Well the word “lockdown” isn’t in the Freedom Fighters dictionary. Same goes for the words “tracking” and “tracing” but “herd mentality” that one is in the Freedom Fighters dictionary. They would rather purposefully risk the lives of others as well as their own lives. Jezz like watching all those rednecks rationalizing why they are at the beach. Dumb and Dumber meets Dead and Deader. We cannot unring those dumbbells.

    We already know what works and what does not work. All I know is that when a vaccine becomes available, that those same Freedom Fighters will find some way to elbow their way to the front of the line. What a bunch of selfish heathens. :/

  78. jacksmith4tx says:

    Everett,
    “All I know is that when a vaccine becomes available, that those same Freedom Fighters will find some way to elbow their way to the front of the line.”
    This will NOT happen. We will be lucky to hit 50% inoculation rates if it’s free. Seasonal vaccinations for children and at-risk groups are down over 20% and we were struggling to just reach levels that were common a decade ago.

    If I could lure all those Freedom Fighters to let me inject them with something I’m afraid would be tempted to OD them all on LSD.

  79. Steven Mosher says:

    Just wait for the clusterfuck

    https://www.northjersey.com/story/news/coronavirus/2020/11/09/nj-covid-vaccine-heres-how-many-get-it-who-gets-priority/6223524002/

    NJ is typical of the state plans I have seen.

    The first people to be protected will be health workers. Of course this makes sense from a couple
    of perspectives but expect to hear things like this
    A) HEY, those people have PPE!
    B) hey those are the people who were never out of a job, they wanted covid to be bad
    C) That’s the health lobby
    D) Hey they are all YOUNG, give it to the old first.
    E) hey some of them already had covid
    blah blah blah

    Then of course some MAGA will go in public, lie, and say they were vaccinated and don’t need a mask. I can see the videos now. Some folks will lie “hey I don’t need a mask, I was vaccinated”
    Can you imagine the chaos when 30% of the population is vaccinated? will they wear mask
    for the 28 days it takes for the “jabs” to take effect. I predict cluster fuck

    If old people get the vaccine First. Expect some to die waiting for their second shot. WHY?
    because old people die. Here in Korea ~100 people died after getting the flu shot.
    NOT RELATED, they just died, cause old people do that. the death rate was totally
    normal, but everyone believes it was from a “bad” flu shot. So now, even here, folks
    are not taking a free flu shot.

    so there are like 25Millon old people and between 500 and 1000 of them die every day
    Chances are as we vaccinate these folks some will die. Because old people do that.
    Now this won’t be from the vaccine!!, but try to explain that to people. Try to explain that
    to red staters who die from Biden’s vaccine. ( kidding)

  80. Steven Mosher says:

    I reserve the right to name call

  81. Jim Hunt says:

    Steven – What name would you choose to call Prof. Levitt?

  82. Joshua says:

    A higher % of people in red states will die after getting a vaccine (there are poorer health outcomes in red states).

  83. Joshua says:

    Levitt actually said that along with the spike in cases in Europe they’re are “no deaths.”

    Michael Levitt’s predictions: for Brazil (he said 98k deaths, currently at 162k,), Iran (he said it was past the halfway mark at 24k cases, it’s at 700k cases now), Switzerland (he said 250 deaths, now at 2,943) Israel (he said likely no more than 10 deaths, currently at 2,681), Italy (he said 17k-20k, it’s at 41,750) and the US (said the pandemic was over on August 22)

    And check out his comments about vaccines.

    But I like his appeal to authority with the Nobel.

  84. I didn’t realise that Levitt was born in South Africa 😦

    I do like that he explicitly says “I’m not a conspiracy theory person at all, but …..” 😀

    It is slightly bizarre that someone with his credentials can really think that working on a topic 10 hours a day for months somehow implies suitable expertise. The overlap between how he presents himself and his expertise, and what I’ve some “skeptics” present, is despressingly similar.

  85. Joshua says:

    I do like that he explicitly says “I’m not a conspiracy theory person at all, but …..” 😀

    https://mobile.twitter.com/MLevitt_NP2013/status/1305791799492071425?s=20

    -snip-

    ML: If Sweden stops at about 5,000 or 6,000 deaths, we will know that they’ve reached herd immunity, and we didn’t need to do any kind of lockdown. My own feeling is that it will probably stop because of herd immunity. COVID is serious, it’s at least a serious flu. But it’s not going to destroy humanity as people thought.

    May 4th:

    -snip-
    TSD: What is your prediction for when the rest of the world will reach a peak number of deaths?

    ML: Something between eight and 14 weeks. But even places like South Korea that have peaked, they’re still seeing deaths. Trying to predict the final death is still very hard.

  86. Joshua says:

    Why does that particular Twitter link only post as the url, and not the tweet itself?

  87. Joshua says:

  88. Joshua says:

    Sorry – Willard told me before… It was the “mobile.”

  89. Steven Mosher says:

    “Steven – What name would you choose to call Prof. Levitt?”

    dumbfuck.

    On twitter I explained to him that CDC excess deaths was provisional. he went ahead
    and used it predicting an end to the pandemic in the USA at 170K deaths, a week or so
    later the excess death data changed under his feet. His prediction was wrong.
    he said ” I was wrong, I’ll get back to it later” Both he and Ioannidis both persist in using
    these gompertz like models to make predictions when everyone in the field knows
    that “Farr’s law” is non mechanistic and not very useful for prediction.
    ( why would be an interesting discussion).

    anyway

    They have a new metric “years lost”. so because old people die, the pandemic is not that
    bad. Ioannidis even uses this to analyze the 1918 pandemic which killed young people
    so “WAY” bad if you look at years lost.

  90. Steven Mosher says:

    “A higher % of people in red states will die after getting a vaccine (there are poorer health outcomes in red states).”

    at one point Levitt was doing calculations on how many old people die after getting a positive PCR,
    but don’t die from Covid. Sorta like if you find 1000 old farts who are positive 10 will die regardless.

    I just see that as they start to vaccinate the old there will be UNRELATED deaths as people are waiting for their second jab. And folks will flip out and the press may flip out and you are left
    explaining to conspiracy nuts that the death was unrelated. Is joe bag of donuts gunna understand that? nope. he’s gonna know that grandpa Fester took the jab and died three weeks later.

    china is facing this now in brazil where a person in the vaccine trial died. unrelated, but
    you have to read past the headline.

  91. Joshua says:

    They’ve been doing the “years of life lost” argument for quite a while now.

    As with the “herd immunity” argument or the “poor children will die because of expensive energy” argument or the “no one will accept slower growth” argument, or the “there were fraudulent votes cast” argument, there’s always a kernel of truth.

    As long as you don’t look at the actual counterfactual nature of your conclusions, don’t factor in externalities, don’t evaluate whether children really will get cheap energy if you don’t subsidize solar, don’t consider whether some people place a higher value on a society’s elders, don’t consider that Sweden is SOSHLIST and effectively has death panels and a really low average # of people per household and relatively few multi-generational households. As long as you’re sure that polls showing Biden with a big lead will suppress Trump votes but not Biden votes. As long as you just roll with correlation=causation, then a kernel of truth suffices.

  92. Actually, the metric should be QALY, quality years of life. Which is tougher to calculate but more relevant.

    Geez. ‘Grandpa Fester?’ Where do you come up with these names, Steve?

  93. Joshua says:

    Looking past the reams of nonsense…

    Mention of military resources being used, and rapid testing?

    https://www.dailymail.co.uk/news/article-8925427/amp/Official-data-exaggerating-risk-Covid-500-academics-tell-Boris-Johnson.html

  94. Dave_Geologist says:

    Actually, the metric should be QALY, quality years of life. Which is tougher to calculate but more relevant

    Already been done Tom. About ten years lost for the median Covid death, and five QALYs. Interestingly, eleven and three* for normal deaths in that cohort. So it’s not just taking the weakest and frailest of the cohort, it’s killing those who were doing very well thank you. Probably including a lot of people with near ten QALYs to balance out those with one or two who’ll still have been disproportionately hit. Stay at Home, Protect the National Health Service, Save Lives: A cost benefit analysis of the lockdown in the United Kingdom. Peer reviewed too, a relative rarity with all this fast Covid publication but I’ll post a comment about that tomorrow (spoiler: wonder how easy the journal found it to review the economics).

    The average age at death and life expectancy loss for non‐COVID‐19 was 79.1 and 11.4 years, respectively, while COVID‐19 were 80.4 and 10.1 years; including adjustments for life‐shortening comorbidities and quality of life plausibly reduces this to around 5 QALYs lost for each COVID‐19 death.

    The fallacy that they were eighty so were due to die anyway is the same as the one which means people can’t understand how the median age of death was 30 in the Bronze age and yet the Bible talks of three score and ten. The average was dragged down by all the childhood and childbirth deaths.

    Note that this is a paper that argues against lockdown on financial grounds, on the basis that the financial costs of lockdown exceed the value NICE places on a QALY.

    How do they achieve that feat? By low-balling the lives saved of course.

    The low cost of effective forms of behavioural change (washing hands, avoiding crowds) adopted by individuals makes it unlikely that in the UK there would have seen 500 000 deaths even with no government restrictions; the 500 000 figure from Ferguson et al (2020) was based on an assumption of no change in individual behaviour. [but it hadn’t brought R below one so we’d just have taken longer to kill our first 50,000 with a longer doubling time; and we’d not have stopped killing them then like we did in the summer].

    The evidence of a turn in the curve before lockdowns are likely to have had much effect is disputed but not easily dismissed. [hmmm, maybe there was herd immunity (hasn’t aged well) … actually there was a slight change in behaviour, but see above].

    Even if lockdowns stopped such huge numbers of deaths over the period March-June 2020 they have not permanently stopped if wider immunity has not significantly risen so that any substantial easing of restrictions will just bring them back. [or, hey, you know, maybe there will be a vaccine: wonder if there’s any news on that front?].

    In many countries deaths were concentrated in care homes for the elderly and have been disproportionately among older people so a blanket lockdown (“don’t leave home”) may have been inefficient—it generated huge costs (see below) and may have yielded limited health benefits, over and above what might have been achieved with measures which focused on groups most at risk. [you mean like Sweden did? – that hasn’t aged well either; and its economic hit is within 1-2% of Norway’s for ten times the deaths per head].

    * That’s not in the paper but I’ve seen it in rebuttals to “but they had ten years left”.

  95. Steven Mosher says:

    “Where do you come up with these names, Steve?”

    But I’m a Munster and Leave it to beaver fan

    haha

    ask me why

  96. Chebyshev says:

    It is amusing to see intelligent folks make the mistake of throwing around technical jargon from a field that is not their own and make inferences confidently. For example, one of the commonly used term in Covid discourse is “case”. Cases exploding, resurgence of cases, etc. There is a world of a difference from a positive test and a case in medical science.

    While covid denialism is a cheap and lazy way of avoiding tough questions and genuine enquiry, I wonder if the term “covidiot” might be useful to capture the group of people who reduce a complex, multi-dimensional problem into a very simple problem of one or two variables that is driven by “legibility”.

  97. @Joshua
    Re: “Nic thus far had been silent on these questions for quite a stretch.”

    Well, Nic Lewis has been wrong over and over on the pandemic, so that isn’t surprising. Politically-motivated epistemic trespassing, and willfully undermining public trust in experts who know than you, often turns out badly.

    Lewis says:
    “it’s good to see that some politicians have a decent understanding of these important issues, and terrifying to see how poor the understanding of them is by supposed scientific experts who control or influence polic policy, and how unwilling such ‘experts’ are to change their views as the evidence against them builds.”
    https://judithcurry.com/2020/09/22/herd-immunity-to-covid-19-and-pre-existing-immune-responses/#comment-927786

  98. KiwiGriff says:

    Quote.
    As we move into the spring/summer period where flu is always uncommon in New Zealand, Professor Michael Baker offers his analysis on the flu season numbers and why masks continue to be so important.

    He said there has been “near extinction of influenza in New Zealand following our very effective Covid-19 response”, as numbers vanished from the two standard systems for surveillance – resulting in a 99.8 percent reduction in flu cases.

    According to Baker, there were usually 1600 more deaths in winter, compared to other seasons, and around a third of those were caused by influenza, mostly in older people with long-term health conditions.

    “What the Covid-19 response has done has largely eliminated those excess winter deaths and mortality as a whole is down around 5 percent,” he said. “So that means an extra 1500 people will survive this year who wouldn’t have.”

    Baker said these measures had led to “a revolutionary change in thinking about how to deal with respiratory pathogens” and could be brought back in the event of a serious flu pandemic.

    “These are not measures you would roll out routinely of course, but if we had a particularly severe respiratory disease like a severe flu pandemic … and it had the same infection fatality risk as we’ve seen with Covid, that is half a percent or one percent of people dying, then we could think about using these measures again.”
    https://www.rnz.co.nz/national/programmes/sunday/audio/2018767843/near-extinction-of-influenza-in-nz-as-numbers-drop-due-to-lockdown

    The effect on life expectancy will be interesting once we have meaningful numbers.
    I predict NZ life expectancy will rise relative to nations that failed to contain the virus to a greater degree than official statistics for covid deaths would suggest .
    FWIW
    As NZ had been brought up.
    We have normal life here in NZ, few even bother to wear masks because we do not have community spread here.
    Most are happy to except occasional regional lock downs that stop any ongoing risk for the majority of us .

  99. Dave_Geologist says:

    Thanks Atomsk’s Sanakan, I had the paper but had missed the SM. From the thread:

    has Curry changed her career path to epidemiology?

    No, she’s still following her old career path, politics and culture wars.

  100. Dave_Geologist says:

    Returning to the QALYs/NICE paper, I wonder what an economists and a thinktanker with an unspecified BSc whose background is as “a technical / operations director who has worked in diverse cultures and industries developing and delivering performance change and improvement through new ideas and people” are doing publishing in a medical journal. In fairness he’s published lifestyle-association studies with the third author, an endocrinologist, on diabetes. But that’s not epidemiology so may be a case of a little knowledge being a dangerous thing. And I think “Water pumps, not Wars” – From emotive to rational language in managing the Covid-19 pandemic belongs in a previous thread, along with self-appointed titles like Professor of Evidence-Based Medicine.

    Regardless, I wonder how well a medical journal was able to review and properly interrogate the economic content and counter-factuals? Their counter-factual seems to implicitly rest on the idea of infection-acquired herd immunity. Otherwise, as I said before, with R still above 1 infection would sweep through the population, with exponential growth until it ran out of victims and moved onto a logistic curve. Killing just as many people in the process as Ferguson said it would (the ONS’s IFR of 0.9% times 70% of the UK population is within a few percent of 500k). Actually it would have been higher, because that was with shielding of the most vulnerable (but I’ll give them that one because they proposed shielding … but forever? … for a year?). I’ve seen estimates of the vulnerable category as high as 30% due to our obesity epidemic adding to age. Go to Euromomo, scroll down to England and filter on 75-84. But then try 65-74 and 45-64. Scary. That’s a lot of the population vulnerable, or at least dying so we can infer vulnerability.

    More importantly, we know that although we had spare ventilators in the Nightingale hospitals, we didn’t have the staff for them. The busiest ICUs were relaxing their rule of one nurse per patient to one per six patients; and in this wave some are already at two. So it would have been Lombardy plus with people triaged on trolleys and a higher fatality rate than 0.9%. And to crown it all, with antibodies fading after four months and with infection only giving 6-12 months immunity in other coronaviruses (common cold), we’d do it again next year until all the vulnerable people are dead. Then it would only kill people as they moved up the age cohort and we’d learn to accept a decade less life expectancy. Basically back to the biblical three score and ten.

    And of course, would deaths in the hundreds of thousands let the economy carry on as normal? Would people go to the pub, visit busy shops, send their kids to school, go to work and drive buses and man checkouts, buy a new car with zero hours most weeks or redundancy threatening? Would tourists fly to a UK that had by a long way the world’s worst death toll? Would they let us fly to them? That’s probably the biggest counter-factual. You can only assume zero or limited damage from not locking down if you think the death toll would not have been an order of magnitude worse, which means consciously or subconsciously you’re thinking that really it’s no worse than flu and the IFR is exaggerated, or that we’re well on the way to naturally acquired herd immunity.

  101. Dave_Geologist says:

    Going back to my claim that the voluntary measures we took in early March were not enough to get R below 1: we know it wasn’t because we have a real-world experiment going on right now. We now know that England’s Tier 3 was enough to maybe-just-about flatten it so R was near 1 but not below 1, and those measures were much, much tougher than anything we’d been doing or would have done voluntarily. And of course if we had implemented them voluntarily there would have been a huge economic impact, lockdown or no lockdown.

  102. Willard says:

    > While covid denialism is a cheap and lazy way of avoiding tough questions and genuine enquiry, I wonder if

    Of course you do:

    New York Time? Hell, no.

    https://judithcurry.com/2020/11/03/u-s-election-discussion-thread/#comment-931765

  103. Joshua says:

    Chebyshev –

    > There is a world of a difference from a positive test and a case in medical science.

    Could you explain what the difference is?

  104. Joshua says:

    Dave –

    > And of course, would deaths in the hundreds of thousands let the economy carry on as normal?

    This is the problem with counterfactuals that I have yet to see any “herd immunity” proponent even attempt to address. Seems such a basic problem to me.

  105. Dave_Geologist says:

    That’s why they also need “nearly there, Joshua”. Which means they need an order of magnitude lower IFR as well. And Dark Matter or Magic Beans immunity.

    Missing the economic impact of massive death, roughly a doubling of our annual deaths from all causes, out of their analysis is a sure “tell” that they believe that, or want so hard to believe it they’re prepared to play Russian Roulette with peoples’ lives.

  106. @Chebyshev says:
    “It is amusing to see intelligent folks make the mistake of throwing around technical jargon from a field that is not their own and make inferences confidently. For example, one of the commonly used term in Covid discourse is “case”. Cases exploding, resurgence of cases, etc. There is a world of a difference from a positive test and a case in medical science.”

    This is my field, so I feel fine about commenting on this again. SARS-CoV-2 is the virus, and COVID-19 is the disease caused by SARS-CoV-2. There’s no problem in medical science with referring to a positive test as a case, as long as one recognize one is dealing with case of SARS-CoV-2 infection. Being a case of SARS-CoV-2 infection is not the same as being a case of COVID-19, just as being a case of HIV infection is not the same as being a case of AIDS. And yes, we do use PCR in medical science to determine if someone is infected with a pathogen, including pathogens other than SARS-CoV-2.

    If you have a problem with “case” being used to refer to a positive test, then I suggest you read the medical literature and familiarize yourself with the “technical jargon” you’re talking about. This is nothing new.

  107. Steven Mosher says:

    I reserve the right to name call

    https://harbingersdaily.com/biden-coronavirus-adviser-vaccine-should-be-distributed-globally-not-to-america-first/

    Like I told you guys when I first started reading these “plans” folks had for distribution
    It will be a cluster fuck

  108. Joshua says:

    I am no fan of Emmanuel (as an aside I know people who know him and who don’t like him)…

    But how would your plan differ? Should a distribution plan not be organized by rate of spread around the globe? Should people in wealthier countries be prioritized?

    And seriously, are you going to go all DEATH PANELS!!!!1!1! on us?

  109. jacksmith4tx says:

    Steven,
    And it’s free* too!
    *proof of a plot to kill MAGA.

    By February there will be multiple vaccines around the world so the roll out will be uneven no doubt.
    The first US vaccine is going to have a pretty complicated distribution challenge too.

    Russia says its Sputnik V COVID-19 vaccine is 92% effective.
    https://www.reuters.com/article/instant-article/idUKKBN27R0Z6

  110. Joshua says:

    Atomsk –

    Obviously, the number of infections would outnumber true cases, but as I understand it a positive test would = a case. Is thar not correct?

  111. Everett F Sargent says:

    “Russia says its Sputnik V COVID-19 vaccine is 92% effective.”
    Yeah, 92% effective in COVID-19 transmission for all social encounters …

    LHS = daily confirmed, RHS = daily deaths
    top = dailies
    middle – dailies as 7-day centered rolling average
    bottom = doubling time in days
    from ~140 days to ~60 days (confirmed)
    from ~120 days to ~60 days (deaths)

  112. Steven Mosher says:

    “But how would your plan differ? Should a distribution plan not be organized by rate of spread around the globe? Should people in wealthier countries be prioritized?

    And seriously, are you going to go all DEATH PANELS!!!!1!1! on us?

    “But how would your plan differ? I would not use Rt to decide. Its not practical.
    Should people in wealthier countries be prioritized? Nope.

    First concerns are practicality , speed, transparency, preventable death, simplicity.

    I think you see a lot of analogs with the green new deal, where people are using one crisis
    to achieve other goals. So, they see poverty as a problem, and they take the vaccine distribution
    problem as an opportunity to address poverty issues.

    Simple question: how will tell people in NYC who have suffered greatly for months,
    from deaths, from riots, from murders, from a destroyed economy, that
    the country of Turkey will get “their” vaccines because Turkey has worse Wealth gap?
    or that China should get their vaccines because China has the WORST wealth gap.

    As much as I would like to solve global poverty, using a vaccine to “cure” poverty
    is just social justice version of HCQ

  113. Steven Mosher says:

    Joshua here is one of their principles
    “It should not be backward looking, punishing or rewarding countries for their COVID-19 response or aiming to redress past injustices”

    How would this work with climate change or reparations?

  114. anoilman says:

    Joshua: I can help with this one….
    > There is a world of a difference from a positive test and a case in medical science.

    Could you explain what the difference is?

    In medicine… there’s what the doctors think and do.. and in medical statistics there is a definition by which you define something. There is a lot of confusion about the two things. One is a measurement used for stats (complete with some ability to measure error), and the other is used for treatment. Doctors like to say what they want, but the epidemiologists just want numbers of known quality.

    Data quality can be a real issue. In your imagination:. a patient presents a positive test result from one hospital.. Who calls that and says what about it? That patient now goes to a different hospital. Who calls that? When is that considered a new case or the previous one? Was it double called? Would a re-infection case be called, when and how? (Is there a data field for re-infection measurement? Nope.)

    US hospital data systems are a bit of a mess. Data systems there are generally geared to singular hospitals. Doctors make all calls on cases, and those are directly used for statistics with zero data cleaning. Are Doctors gonna let epidemiologists change their patient records? nope. (This is how its done in Epic: https://www.epic.com/software)

  115. Joshua says:

    Oilman –

    Sure, there are issues with data quality. But as far as I know, a “case,” or a “true case,” is basically identified as a positive test. Hence the difference between the case fatality rate and the infection fatality rate.

    I would appreciate knowing if I’m wrong about that.

  116. Joshua says:

    Just because Matt is a friend of the blog :

  117. Dave_Geologist says:

    Steven, I reserve the right to name call.

    Setting aside the question of whether someone advocating ethical distribution is a good, a bad or a naive person.

    Smear-by-association is not a good look.

  118. Joshua says:
    “Obviously, the number of infections would outnumber true cases, but as I understand it a positive test would = a case. Is thar not correct?”

    Yes, a positive test is a case of SARS-CoV-2 infection. And yes, the total number of infections is larger than the number of reported cases. Your understanding is correct.

    In other news:

  119. Joshua says:

    Thanks Atomsk.

    It’s notable how commonly were see the misunderstanding that the death rate can be decoupled from the infection rate. Nic Lewis and the rest of the “herd immunity” crowd compare infection rate to the contemporaneous death rate to reach a conclusion that they CAN be decoupled. Look at Nic’s comments from 8 weeks ago and you can see that misperception (e.g., that an increase 8 weeks ago might just be an isolated “blip,” or the observation that an increase in cases wasn’t associated with a rise in hospitalizations over a relatively short lag, or the suggestion that the increase in cases)

    Apparently the reality that you have to allow for a significant lag between positive case identification and hospitalizations, ICU admissions, and deaths is a difficult concept for them to grasp. They’d rather engage in magical thinking about “T-cell immunity” or miraculous treatment improvements or the notion that massive increases in infections among young people won’t transfer to increases in older people or thar the increases in positive tests are only caused by increased testing (irrespective if what we see reflected by positivity rates).

    We’ve seen this magical thinking over and over.

    One might be tempted to generalize about why it is so prevalent at a website where the “denizens” largely identify as “skeptics.”. It doesn’t seem like skeptical thinking at all to me.

  120. Joshua says:

    Useful for evaluating the “herd immunity” arguments (of course, relative # of tests should be considered)

  121. @Joshua said:
    Re: “Useful for evaluating the “herd immunity” arguments (of course, relative # of tests should be considered)”

    I’m not surprised. I’ve been pointing out for months that Nic Lewis and others aren’t right about the herd immunity threshold being low, to the point that Judith Curry started blocking my comments to shield Lewis from criticism. It’s literally gotten to the point that a non-expert can post dangerous misinformation in my field of expertise, top USA government officials will re-tweet that obvious misinformation, and I’ll get blocked for correcting them. This is wild. It’s reminding me of what happened with Mbeki’s government in South Africa during the HIV/AIDS pandemic. Lewis owes folks an apology.

    Me, on May 15 in response to Lewis’ initial article on the herd immunity threshold (HIT):

    “So it’d be great if you were right about HIT being low; I don’t want to see more people die and lockdowns are painful for many people. Also, it would be politically-convenient for a lot of right-wing people if you were correct, just as it would be convenient for ECS to be 0.6K so they can avoid climate policies they dislike. But there’s every reason to think you’re wrong, just as there’s every reason to think ECS is not as low as 0.6K.”
    https://judithcurry.com/2020/05/10/why-herd-immunity-to-covid-19-is-reached-much-earlier-than-thought/#comment-917050

  122. Chebyshev says:

    Positive test vs cases:

    1. Positive test could mean “dead virus particles”. In fact, it appears it is the case for a significant %: https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/10.1001/jamainternmed.2020.7570
    2. False positives.
    3. Asymptomatic people.
    4. If you test positive and self quarantining, watching TV and sipping beer and are not in a clinic, you are not a case. I am sorry.

  123. Steven Mosher says:

    “4. If you test positive and self quarantining, watching TV and sipping beer and are not in a clinic, you are not a case. I am sorry.”

    wrong.

    You are a case as far as epidemiology is concerned.
    You are case f as far as the contact tracers are concerned
    You are a potential case for the guy who has to plan for open beds in the hospital because
    you may take up a bed in the future.
    You are case as far as people in your house are concerned, because they should be cautious around you.

    you are not a case for the doctor or mortician.

    yet.

    here is the thing. you don’t get to walk into a field or subject matter and tell people how they should use words, any more that I can tell a Brit that they misuse the word rubber

  124. Everett F Sargent says:

    Chebyshev,

    Wrong answers on all four counts. Please provide direct authoritative links defining so-called cases, not your own personal opinion on what you think are cases. The giveaway? Using so-called ‘False positives/ as a fictitious item in your fictitious list.

    It would be nice to use hospitalizations as the definition of actual cases as you appear to be defining cases, as that would lead to very high CFR.

    Don’t hide behind a screen name, if you actually are a Subject Matter Expert (SME) at the level of say Anthony Fauci, ha ha ha.

    Oh and you are the interboobnets #1 and #2 effin’ fools.

  125. Joshua says:

    Chebyshev –

    > 1. Positive test could mean “dead virus particles”. In fact, it appears it is the case for a significant %:

    My guess is that this is not news to anyone here. We have discussed this before in the context of implementing rapid (antigen) testing on a widespread basis.

    I’m guessing that you are pointing to this as a reflection of the “casedemic” logic. As per usual, there’s a kernel of truth to these banal arguments from “skeptics.”

    But the rise in hospitalizations, ICU admissions, and deaths is the problem with the “casedemic” arguments. What we see are foolish people looking at contemporaneous trends and not allowing for the lags that we know exist. If you need more explanation there let me kniw and I’ll be happy to provide it.

    > “4. If you test positive and self quarantining, watching TV and sipping beer and are not in a clinic, you are not a case. I am sorry.”

    Yeah. I thought that’s were you were going. Thanks for confirming my intuition.

    In point of fact, a positive test is a “case.”. As in determining a case fatality rate

  126. Everett F Sargent says:

    Challenges in Testing for SARS-CoV-2 Among Patients Who Recovered From COVID-19
    (the title is all you really need to read, do’h, the sample had 100% COVID-19)
    https://jamanetwork.com/journals/jama/fullarticle/10.1001/jamainternmed.2020.7575?utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamainternmed.2020.7570

    Among patients who have recovered from COVID-19, repeated testing for SARS-CoV-2 may be done weeks or months after infection either as part of routine screening (eg, screening nursing home personnel on a weekly basis to prevent transmission of infections to patients) or because of the development of symptoms that are worrisome for reinfection. Unfortunately, the interpretation of positive test results in patients who have previously recovered from COVID-19 is fraught. The best widely available test, a real-time polymerase chain reaction (RT-PCR), is very sensitive for fragments of viral RNA and can be positive because of nonviable remnants of the virus. Currently, there is not a widely available test for determining whether the virus can reproduce and transmit infection.

    In this issue of JAMA Internal Medicine, Liotti et al1 describe the results of retesting 176 patients who had recovered from COVID-19 with 2 negative RT-PCR test results 24 hours apart. At a mean of 48.6 days from their date of diagnosis, 32 patients (18.2%) had a positive PCR test result for SARS-CoV-2 RNA. Using a specialized assay, only 1 of these 32 patients (3.1%) had evidence of RNA capable of replication. Although this study cannot solve the challenge of interpreting positive PCR results in recovered patients, the data help us to better understand the scope of the problem.

    To avoid unnecessary quarantine for patients who have recovered from COVID-19, routine repeated PCR testing should not be done in the 90 days following infection. However, more complicated is what to do about patients who are symptomatic and have positive results on repeated PCR tests. Reinfection with SARS-CoV-2 has been documented2 (based on demonstration of different genetic differences between the viruses infecting the person on the first and second episode) but is rare. Until clinical laboratories have the capability to test for the reproductive capacity of coronavirus, interpretation of the epidemiologic significance of positive PCR results among recovered patients will remain challenging.

  127. Everett F Sargent says:

    The Need for More and Better Testing for COVID-19
    https://jamanetwork.com/journals/jama/fullarticle/2773129

    I read it and I think it deserves a read by otters.

    “Even as efforts continue to address these public health testing challenges, recent highly publicized outbreaks, such as university campuses and the White House, are a stark reminder that testing alone is also not sufficient to prevent community transmission. It is more accurate to consider testing less of a prevention strategy than a mitigation strategy. Testing in the absence of other proven prevention strategies is unable to prevent outbreaks.10 Even as tests become faster with higher sensitivity and specificity, social distancing, mask wearing, and avoidance of large indoor and outdoor gatherings must remain central to any public health strategy. Although the evidence is growing that widespread access to rapid antigen testing may be a pragmatic tool to interrupt the community transmission of SARS-CoV-2, what will remain equally important to prevent spread of infection to others is what happens before and after test results are delivered. Even the perfect test cannot go it alone.”

  128. Steven Mosher says:

  129. @Chebyshev says:
    “Positive test vs cases:
    1. Positive test could mean “dead virus particles”. In fact, it appears it is the case for a significant %: https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/10.1001/jamainternmed.2020.7570
    2. False positives.
    3. Asymptomatic people.
    4. If you test positive and self quarantining, watching TV and sipping beer and are not in a clinic, you are not a case. I am sorry.”

    I can typically recognize the ideologically-motivated non-experts who never heard about PCR before the pandemic, because they repeat the misinformation you just did. It’s the intellectual equivalent of flat-Earthers saying “I see no curve”.

    SARS-CoV-2 is a virus, and therefore is neither alive nor dead. So the phrase “dead virus particles” is nonsense. Moreover, anyone who has a clue about PCR testing (I do; I’ve run it since high school, starting over a dozen years ago) knows it works by amplifying genetic material, whether that be DNA or RNA (after reverse transcription, in the case of RNA). Its purpose is to tell you whether virus-specific genetic material is there, which would happen with infection. Complaining that it detects “dead virus particles” is as ridiculous as objecting to hammers by saying they won’t help you screw a nail into a board (as if nails were the sorts of things that could even be screwed in, and as if the purpose of hammers is to screw things). It shows you fundamentally didn’t understand PCR, nor its role. So folks, if anyone ever tells you about PCR tests detecting “dead virus particles”, know you’re dealing with someone who is either uninformed and/or willfully misleading you.

    Your second point is irrelevant, since whether a test yields false positives is irrelevant to what qualifies as a case. By that implausible logic, there are no almost no cases of any disease, since virtually all of the diagnostic test/criteria we use in medicine, sometimes yield false positives.

    Your third point still doesn’t acknowledge the difference between a case of infection vs. a case of disease; the former requires no symptoms. And your fourth point is just your non-expert personal opinion on a topic you don’t have expertise in. It doesn’t even make sense when discussing a case of disease, let alone a case of infection, since there are plenty of diseases in which a disease case will be outwardly asymptomatic and sitting at home. For example, there are plenty of type 2 diabetes patients watching TV at home and not at clinic. You don’t need to be in clinic to be a case of disease.

  130. Willard says:

    > I can typically recognize the ideologically-motivated non-experts

    No need to probe minds, Atomsk.

  131. David B Benson says:

    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7158286/
    Virus Life Cycle by Wang-Shick Ryu begins with “Viruses are obligate intracellular parasites.”

    As for parasites, we see in
    https://en.wikipedia.org/wiki/Parasitism
    that such are classified as species.

    So it’s convenient to consider viruses to be living organisms, despite what elementary biology texts often state.

  132. Dave_Geologist says:

    More on PCR vs infectiousness: Duration of infectiousness and correlation with RT-PCR cycle threshold values in cases of COVID-19, England, January to May 2020 .

    Severe acute respiratory syndrome coronavirus 2 viral load in the upper respiratory tract peaks around symptom onset and infectious virus persists for 10 days in mild-to-moderate coronavirus disease (n = 324 samples analysed). RT-PCR cycle threshold (Ct) values correlate strongly with cultivable virus. Probability of culturing virus declines to 8% in samples with Ct > 35 and to 6% 10 days after onset; it is similar in asymptomatic and symptomatic persons. Asymptomatic persons represent a source of transmissible virus.

    I presume questions remain about whether a cultivable amount of virus would transmit a sufficient viral load to actually infect another person (I haven’t heard about it being hard to culture, unlike say Epstein-Barr virus until Barr hit on the method, so presume it would be that way round rather than failing to culture from a sample that was capable of infecting someone). The consistency between the cultivability results and the epidemiologically derived ten days since symptom onset suggests a pretty good match.

    Not being a PCR expert, I don’t know how transportable the exact Ct value would be across laboratories, machines and reagents.

    Strengths include the comparatively large size of this dataset, inclusion of a large proportion (greater than 50%) of samples taken more than 7 days after symptom onset and that all analysis was performed in a single laboratory.

    In the UK more than half of tests are done in the standardised Lighthouse Labs which were set up in the spring.

    One straw in the wind for me comes from the fuss about the UK’s fast lateral flow tests being claimed as 100% effective in detecting cases positive on PCR, and an NHS assessment finding them only 50% effective. They claimed 100% at a Ct of 33: if you eyeball Fig. 2, 33 to 35 looks like about a factor of two in % culture-positive, which would be consistent with 100% to 50%. I wonder if that’s not stacking the deck in favour of the rapid-flow test, but rather an honest answer to a different question: at what Ct does the rapid test perform equally to the PCR test (recognising that there’s no reason for PCR to stop there). IOW not “as good as PCR”, but “as good as 33 cycles of PCR”.

  133. Dave_Geologist says:

    Digressing to the question of whether viruses are alive. I would say yes. Lacking the machinery to replicate without a host is no different from a gazillion parasites, some of them animals which are definitely alive. I presume it’s because they don’t just rely on the host for food and chemicals, they need it for transcription and replication. But why reify that aspect in particular: lots of inanimate objects from the simplest crystal can replicate but are not alive. And what about DNA viruses? Surely they’re not a separate branch of parallel-developed DNA “life”? That would be too much of a coincidence. And in any case would require that in their ancestral form they could replicate without a host, so were truly alive and we get a Catch-22. Much more likely surely that they’re an extreme form of parasitism within the same sort of DNA life as us. And some even have their own means of executing transcription and replication.

    dsDNA viruses can be subdivided between those that replicate in the nucleus, and as such are relatively dependent on host cell machinery for transcription and replication, and those that replicate in the cytoplasm, in which case they have evolved or acquired their own means of executing transcription and replication.

    But I’m fine with viable, especially given its etymological origins 😉

  134. Dave_Geologist says:

    FWIW my Just So Story starts with TNA life. A less efficient replicator than RNA but it can be made from abiotic compounds we know were available because they’re found in meteorites and have been detected spectroscopically in interstellar molecular clouds. Once it’s got going its metabolism stirs enough complex molecules into the primordial soup for the precursors of RNA to be assembled. RNA life outcompetes TNA life, either directly by eating it or indirectly by eating its lunch. Rinse and repeat for DNA, except because unlike TNA life it uses the same chemistry as its successor, RNA life is able to hold out as a parasite after all the self-sustaining RNA life had fallen by the wayside.

    As we’ve discussed in exoplanet threads, the hardest step towards life is the first one, and for me that’s where the living/non-living boundary should be placed.

  135. David B Benson says:

    TNA?

  136. Steven Mosher says:

    “Not being a PCR expert, I don’t know how transportable the exact Ct value would be across laboratories, machines and reagent”

    CT values are complicated.

    Different PCR tests target different genes, the WHO standard ( and korean tests) target
    3 genes, N, E, and RdRP. In some tests each of the genes will have a different CT value
    ( be detected at a different thermal cycle) In other tests they may only target 2 genes ( use 2 primers). In some machines like the Abbott the machine will ADD 10 to the reported CT
    so a 24 will report as 34.

    In some cases mutations in the N gene have caused false negatives, cases in Germany I recall

    the process is not designed to be a QUANTITATIVE result, although the process is capable of
    reporting CT numbers.. or the median of CT, or the mean of CT, or CT values for each gene.

  137. Steven Mosher says:

    “at what Ct does the rapid test perform equally to the PCR test”

    perform equally doing what?

    1. testing for “viable” virus? no test does that. you have to try to culture the virus in a level 3
    takes a couple days.
    2. test for presence of genes?

  138. Dave_Geologist says:

    Steven: #2 (or fragments of genes, or the protein or whatever else the particular test targets).

    IOW if I had a patient who gave a positive PCR test after 33 cycles, the lateral flow test would also give a positive result for that same patient so not miss anyone. It would however miss people at the Ct number the Lighthouse Labs are routinely using, and also give many more false positives (summer tests with low virus prevalence were returning 99.8-99.9% negative, so the PCR false positive is very low as it must be less than or equal to the number of positive tests). I understand that in Liverpool positives were to be confirmed with PCR, isolating while they waited to see if it came back negative. During the rising part of the second wave both should detect relatively low proportions of relict, non-viable virus components from a previous infection. That will be more of a concern in the falling part, but better to stop someone who’s clear seeing granny than to have someone who’s not clear infecting granny.

  139. Dave_Geologist says:

    TNA David: threose nucleic acid. It can interact with DNA and RNA but is resistant to nucleases, hence my speculation that as RNA life arose it out-competed TNA life for resources rather than eating it.

    There’s also a candidate for an earlier stage in PNA (peptide nucleic acid), which has additional advantages for abiotic synthesis. How life began on Earth: a status report.

    Both have been synthesised because they’re of interest to the pharmaceutical industry.

  140. Steven Mosher says:

    I think its covered here

    https://www.nejm.org/doi/full/10.1056/NEJMp2025631

    lateral flow has a LOD 100 to 1000 times less sensitive than PCR.

  141. Dave_Geologist says:

    Thanks Steven, useful. The Scottish Clinical Director got into hot water in the papers when he was asked about rolling out millions of PCR tests for mass population testing and said it was not fit for purpose. He mean for that purpose, but of course that was accidentally or deliberately missed by journalists and commentators.

    When someone can be asymptomatic or presymptomatic but infectious before even PCR detection the trade-off in mass testing favours speed and convenience over precision. Better to test someone twice and catch them half-way through asymptomatic infection than to test them once a day too early. Also, and he was probably not prepared to be so blunt on air, it’s not necessary to get R to zero to reverse the growth in cases. If R is close to one and you only catch half of the asymptomatic carriers, you’ve knocked R below 1 and the spread will reverse. More people will die than if you got it to zero, but a lot fewer than if you’d left R above 1 because it was just impossible to carry out such a testing programme using PCR. And with a three day reporting time, even those who test positive with PCR will have been on the streets, infectious, for days.

  142. Ben McMillan says:

    If these lateral-flow tests really do have a 50% false negative rate (versus gold-standard and in community settings), then it will be interesting to see how they get ‘marketed’. Public health in other bits of the UK are already looking at rolling them out.

    People want them as ‘safe to go about my business’ tools but they are much better as sentinel tools in conjunction with test+trace. Catching half the cases in an area would make a huge difference if most will isolate, and you can target their contacts. You could catch most clusters.

  143. Joshua says:

    > Better to test someone twice and catch them half-way through asymptomatic infection than to test them once a day too early.

    I’ve tried to explain to people that repeating rapid antigen tests diminishes the chances of a false negative to be vanishingly small and can be more useful than an expensive test that is more accurate (per test) but given infrequently… but it’s a hard sell.

    Often I get “Well, how well did it work at the Whitehouse?” – even though (1) exceptions will happen and (2) we have no idea what actually happened at the Whitehouse – there’s prolly a reason they wouldn’t say when they conducted testing and when they got negative test results.

  144. Joshua says:

    > I’ve tried to explain to people that repeating rapid antigen tests diminishes the chances of a false negative to be vanishingly small…

    Of course, assuming each test result is effectively independent – which I guess is at least somewhat questionable.

  145. Steven Mosher says:

    “If these lateral-flow tests really do have a 50% false negative rate (versus gold-standard and in community settings), then it will be interesting to see how they get ‘marketed’. Public health in other bits of the UK are already looking at rolling them out.”

    mina suggests a lateral flow test with two tests. basically testing two different genes.
    negative on gene 1, take the second test.

    the key is catching positive infectious people FAST so you can
    A) isolate them and
    B) get to their contacts

    in Korea they use pcr exclusively, but then we get results in less than 6 hours.

    basically if you test positive the contact tracers have 1-2 days to contact everyone
    their biggest fear is over running the contact tracing capacity and then the ICU beds
    there are , 200 open beds for a nation of 50M.

    the game of wack a mole continues and cases have jumped to 200 a day.
    tighter controls coming and they have warned the public

  146. Steven Mosher says:

    “I’ve tried to explain to people that repeating rapid antigen tests diminishes the chances of a false negative to be vanishingly small and can be more useful than an expensive test that is more accurate (per test) but given infrequently… but it’s a hard sell.”

    ya the safety freaks seemed bugged by false negative and the freedom freaks are bugged by
    false positives.

    there is a similar weird thing here in korea, they make and sell rapid tests, but when I do quarantine here they always use PCR.. on day 1 and day 14? . I look at it and think, damn you have me
    here in quarantine along with 2000 other people why arent you using your rapid test AND the PCR
    to see if the rapid test is any good. I’m a willing subject, I have to agree to whatever they want
    every day the bunny suit dude came to the door to take my temp, heck he could shove more
    sticks up my nose for science and its not like I could complain, heck I want them to use this
    experience for more science.

  147. Dave_Geologist says:

    Josua, I think the biggest risk there is the human factor, especially for self-testing. Did I swab right (did I even try if I’m asymptomatic and on a zero-hours contract with no sick pay)? Did I observe proper hygiene when swabbing myself and my kid? Hence the benefit of a urine of finger-prick test if available. From TV the Liverpool ones seem to be self-swabbed with a staff member telling you how to do it right and watching you do it.

  148. Steven Mosher says:

    “Also, and he was probably not prepared to be so blunt on air, it’s not necessary to get R to zero to reverse the growth in cases. If R is close to one and you only catch half of the asymptomatic carriers, you’ve knocked R below 1 and the spread will reverse.”

    Ya. this is so easy to say the wrong thing and get misunderstood or misrepresented

    Now I will say I am shocked that UK hasn’t handled this as well as I expected them to.
    the USA was always gunna be a basket case, but I expected more of the UK you always seemed to
    be more orderly and rational than us colonists. maybe I lost touch or idealized ya’ll.
    I was also shocked that the aussies handled it so well–compared– I expected them to be total
    lunatics.

    when this is done, what changes do ya’ll think will be made to public health in the UK?
    what lessons will sink in?

    covid is something like a contrast agent, you know that junk they put in your veins to allow them to see stuff better. its a contrast agent for your society, everything hidden becomes visible

  149. Steven Mosher says:

    “Josua, I think the biggest risk there is the human factor, especially for self-testing.”

    ya, I think the best use case is something like
    A) you go to work, nurse comes around on rounds, you lick the paper and go back to work.
    B) kids go to school, nurse comes around, kid licks the paper and goes back to class.

    there are also breathalizer versions in the works.

    damn this thing

  150. Joshua says:

    > ya the safety freaks seemed bugged by false negative and the freedom freaks are bugged by
    false positives.

    It’s not just freakishness, though. I tried to explain to someone who does triage for people calling in to a medical line for covid.

    I said (for an individual that was symptomatic) that they should do a rapid test while waiting for the result of a PCR test – because that would help to minimize the chances of a false negative on the PCR test. I couldn’t get across the idea that a less accurate test could improve on the results of a more accurate test.

    > damn you have me
    here in quarantine along with 2000 other people why arent you using your rapid test AND the PCR
    to see if the rapid test is any good.

    Sheece. Just makes no sense. Well, I take some solace that Americans don’t hold a monopoly on dumbness.

  151. Joshua says:

    Conditional probability is just a bear. And it’s a weak point that motivated reasoning feasts on.

  152. Dave_Geologist says:

    the safety freaks seemed bugged by false negative and the freedom freaks are bugged by false positives.

    Indeed. And I think we’d find many of those opposing masks now were saying back in the spring that we should all wear masks like in Asia where they controlled it. Completely ignoring that it was masks and other control measures, the latter often more draconian and intrusive than anything proposed in the West. Of course that was when the argument was “wear masks and no need for lockdown”.

    Back to residual RNA fragments on PCR: we know from the UK ONS survey data in the summer that they disappear much more rapidly than antibodies. Otherwise we could never have had weeks with PCR positives from random people in the low fractions of 1%, a couple of months after the first wave, but positive antibody tests approaching 10%.

  153. Dave_Geologist says:

    David, further to DNA precursors: The RNA World: molecular cooperation at the origins of life.

    It is generally believed that DNA arose after both RNAs and proteins because the usual biochemical synthesis of deoxyribose is from ribose, and because several non-homologous forms of DNA polymerases are found across the domains of life, whereas the ribosome itself is homologous in all domains

    The non-homologous forms of DNA polymerases might suggests multiple strands 🙂 of DNA life, independently evolved. But the homologous ribosome suggests divergence from a common ancestor. And even if they had independently evolved you still have a Catch 22 in saying DNA viruses are not alive: they must have had a self-replicating ancestor at some point tn the past to get from the soup stage to the virus stage. You then have to say when in its evolution it became no longer alive.

  154. verytallguy says:

    Mosher

    Now I will say I am shocked that UK hasn’t handled this as well as I expected them to.

    You may not have been paying attention Mosh.

    Our current administration is led by a lazy narcissist twice sacked from previous jobs for lying, a man so dishonest he literally refuses to say how many children he has.

    Its raison d’être is to enact a nationalist policy which brought it to power and is predicted by every serious expert to impoverish the country whilst they loudly insist the opposite will happen. They openly deride professionalism saying we are “sick of experts”

    Which part of failing to rapidly enact difficult policies which depend on expert guidance and rational evaluation of evidence are you finding surprising?

  155. Everett F Sargent says:

    vtg,

    I will see your BoJo and raise you with two very Small Hands. 😉

  156. Everett F Sargent says:


    SCC = Americas – (CA +US)
    RoW = World – (CN + EU +US)

    A 14-day lag between deaths and confirmed produces the flattest ratio between their doubling times (from about day 70 to the present), say 1.6 to 2.0 (essentially a lagged autocorrelation).

    The above uses James Annan’s trick of averaging each day of the week to largely remove the weekly reporting cycle modulation on doubling times. No more log-normal plots for doubling times either (log-normal plots tend to hide both the weekly modulation and relative changes when their doubling time values tend towards larger values). Death spikes have been largely removed and folded back per that country’s entire death time series (proportionately). Not reintroducing those death spikes results in a loss ~2% of deaths. Folding that 2% back into the totals does not change the doubling time plots visually. Nothing has been done with the confirmed time series as these appear to be more date accurate (there are spikes but not large enough ones to see in a centered 7-day rolling average).

  157. @Joshua

    Curry’s blocking comments to shield Lewis from criticism. Again. Not surprising, since anyone with even a basic grasp of the the biology and epidemiology on SARS-CoV-2, can see how badly Lewis under-estimated the virus’ fatality rate in various regions.

  158. Steven Mosher says:

    “You may not have been paying attention Mosh.”

    no I have. The thing that shocks me is the public.
    leaders will always fuck up. I’m not shocked at that.
    I kinda expected the UK citizens to do the right things regardless of administrations.

    I see two stupid responses.
    A public that demands the government “do something”, when it basically comes down to the publics
    actual behavior
    A public that demands the government do nothing and thinks they will manage their own risk,
    and then fails to

    And yes, the government can make it better by supporting those who
    modify their behavior. remember 80% of all cases typically come from SSE.
    And yes the government can make it worse,
    but in the end I am more shocked by the public.

  159. Steven Mosher says:

    “Otherwise we could never have had weeks with PCR positives from random people in the low fractions of 1%, a couple of months after the first wave, but positive antibody tests approaching 10%.”

    ya, also the ability of the thing to silently spread seems substantial.
    who knows maybe all “flus” spread silently in the warmer months..
    weather warms.. innoculums decrease, “everyone” gets an asymptomatic case

    it’s a puzzling little critter

  160. Everett F Sargent says:

    Well, changing the window function (to a 7-day rolling re-weighting window) makes the doubling time plots look somewhat better …

    LHS = as is raw data doubling time
    RHS = spike removal (deaths) and re-weighting window applied (deaths/confirmed)

    I need to try this on several country subsets and several of the larger individual countries. I can almost get away with a 3-day OLS doubling time fit now. 🙂

  161. verytallguy says:

    “The thing that shocks me is the public.
    leaders will always fuck up. I’m not shocked at that.
    I kinda expected the UK citizens to do the right things regardless of administrations”

    Haha

    Yes, Great British Common Sense turned out not to be so, did it?

    What can I say? We’re idiots, but I exactly surprised by b that. Interesting what your comments tell us about others’ perceptions of us.

  162. vtg,

    Interesting what your comments tell us about others’ perceptions of us.

    It’s the accent 🙂

  163. Joshua says:

    Perhaps might be of interest…as something of a comparison to climate change modeling and economics.

    First, this article:

    https://amp.economist.com/finance-and-economics/2020/11/14/why-relations-between-economists-and-epidemiologists-have-been-testy

    The a response (and responses to the response):

  164. Willard says:

    In fairness, relations between economists and everyone else have been testy.

    Witness:

  165. Pingback: Namecalling in science – youcancallmeal

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.