Jump to content

CaptainSwing

Members
  • Content Count

    1,779
  • Joined

  • Last visited

About CaptainSwing

  • Rank
    Registered User

Recent Profile Visitors

651 profile views
  1. Well, nobody is 'trashing' it. There's quite correctly plenty of criticism of it, especially here but also here, but the one thing the Oxford group, its critics and I all agree on is that a large scale, randomised, serological survey is the only way of knowing for sure how many people have been infected, which is a piece of information that is crucial to the management of the outbreak. And hopefully it will soon be feasible, if the new antibody tests prove reliable. [For avoidance of doubt, I'm not taking the Oxford report as gospel, or even particularly defending it except as a decent and thought-provoking piece of work.]
  2. Broadly speaking there are two possibilities, both consistent with the progress of the epidemic so far: 1. The percentage of people who develop serious illness when infected (call this 'rho') is relatively big, in which case the percentage of people who've already been exposed to the virus (call this 'z') has to be relatively small, or 2. rho is relatively small, in which case z has to be relatively big. This is what the Oxford group argue, based on established methodology and estimates of rho from the WHO and published in the Lancet, plus circumstantial evidence like the first case in the UK having been recorded back at the end of January. As mentioned several times, you can't easily discriminate between these two hypotheses, at this stage, without doing a proper survey of the population as a whole. Hopefully the blood drop antibody test will be shown to work OK, which would make such a survey feasible.
  3. First two confirmed cases in the UK were on 31st January according to the ECDC dataset.
  4. True, but the Oxford report isn't primarily trying to determine mortality rates.
  5. I'm not really up to speed on this, so this might be a daft question, but would that still apply if there was an antibody test? Presumably the only tests currently available are looking for viral RNA?
  6. Not really speculation. More a testable prediction based on a well-established model and informed by expert opinion and such data as are available so far. Other models, equally plausible a priori, might come up with very different numbers for the fraction of people who've already been infected. The only obvious way to decide between them would be to test a random sample of say one or two thousand people. Also, the case fatality rate in South Korea is currently standing at 1.4% and rising, according to the ECDC numbers collated by Our World In Data. By the way I'm not saying that the Oxford report is flawless. I could pick a few holes in it. It's not been peer reviewed, and is work in progress.
  7. To clarify / repeat, the '0.1% to 1%' figure in the quoted report is the fraction of people who would get seriously ill if infected, which would include (some) old people, people with existing conditions, healthy young people who might suffer a cytokine storm, etc. The figure they use for the proportion of these people who die is about 14%. Multiply those together and you get an even tinier number. The point of the draconian steps is that even with these tiny fractions you can still end up with an Italy-like situation in which the health service is swamped, due I guess to the virus being unusually contagious, so that the outbreak explodes. And even tiny fractions add up to a large absolute number of deaths. Plus there's the consideration of long-term lung damage to some of the people who survive. Your suggestion of isolating vulnerable people until vaccines or treatments become available sounds like a good one. It would have its disadvantages (stigmatization of the isolated and their families, can't be sure who is vulnerable, slowness, administrative difficulty and leakiness of implementation), but as you say the current strategy has its disadvantages too. It's very true that the case fatality rate (no. of deaths divided by no. of confirmed cases) isn't particularly meaningful [that kind of bias is the bane of retrospective studies in particular], but it's all we've got in the absence of (as I say) a proper randomised survey of the population, so it's not surprising that people latch onto it.
  8. A paper (doubtless one out of many) on the effectiveness of different masks, including homemade ones, is under the 'Download full-text PDF' button here: https://www.researchgate.net/publication/258525804_Testing_the_Efficacy_of_Homemade_Masks_Would_They_Protect_in_an_Influenza_Pandemic To me, their conclusions are a lot more pessimistic than their data, which seem to show that even the improvised ones are of some benefit, provided you don't leave gaps round the side. The virus seems to survive for at most a few days on most surfaces: https://www.economist.com/graphic-detail/2020/03/19/how-long-can-the-novel-coronavirus-survive-on-surfaces-and-in-the-air so maybe you could make or buy a small collection of masks and cycle through them?
  9. The FT article is behind a paywall, but the research on which it's presumably based is available for free, by clicking on the dropbox link in the following tweet: https://twitter.com/EEID_oxford/status/1242402762283012096 I can help interpret that if anyone wants. One thing to note is that they apply their model to Italy as well as to the UK. The Italian health service is getting swamped even if it is 'one in a thousand'. On the other hand, using that assumption something like 80% of the Italian population will already have been exposed/infected (grey curve in Figure 2B, which shows the proportion not yet exposed), which would be good news on the herd immunity front. [Their corresponding figure for the UK is 68% by 19th March, but, to repeat, that is based on the assumption that 0.1% of the population will get seriously ill when infected. But that assumption is supported by the fact that if they assume a bigger number, say 1%, they get counterintuitive results like the first cases being reported before the virus got into the country (for the UK, Figure 1E; they don't get that for Italy, Figure 2E).] Is their model right? Time will tell. Or, as they say, a proper randomised survey of the population, using antibody tests, would settle the question more quickly.
  10. Thanks FG, that's really good news.
  11. It doesn't necessarily have to be for months. Given that the disease runs its course in a few weeks, you can get it under control if you maintain near-complete isolation for about that length of time, or a bit longer to be on the safe side. That's what happened in Wuhan. The longer you leave it before you implement the lockdown, and the less complete the lockdown, the worse the outbreak will be - let's hope they haven't already left it too late in the UK, and that the measures go far enough. [In particular, I think, home delivery of essentials needs to be getting organised on a vastly bigger scale, again as happened in Wuhan.] Of course you're not out of the woods after initially getting it under control, you still need to test and track, and further lockdowns might be needed down the line. But at the very least it buys you time to develop vaccines and treatments, as the 'hammer and dance' article I linked to just now argues.
  12. Here's the original article, behind a paywall (so I haven't read it): https://www.thetimes.co.uk/article/coronavirus-ten-days-that-shook-britain-and-changed-the-nation-for-ever-spz6sc9vb The government are vehemently denying it. [Remembering Yes, Minister, does that mean it must be true?] The denial is covered in the following article, from which you can also infer roughly what was said in the first one: https://www.theguardian.com/politics/2020/mar/22/no-10-denies-claim-dominic-cummings-argued-to-let-old-people-die Both articles are (or seem to be) really talking about the sudden change of plan from 'herd immunity' to 'lockdown' following the publication of the Imperial College report - from (something like) 'mitigation' to 'suppression', in the terms of the paper. As I've mentioned before, the 'herd immunity' strategy was always going to involve a very large number of people dying, both directly and as a result of the health services being swamped. There's nothing controversial about that. See the following article from a few days ago, which makes a very strong case for the suppression strategy: https://medium.com/@tomaspueyo/coronavirus-the-hammer-and-the-dance-be9337092b56 [He's actually a bit harsh on the Imperial report, I think - their 'adaptive trigger' strategy is basically the same as his 'hammer and dance', though perhaps they didn't give it enough emphasis. Here is the Imperial report, for reference: https://www.imperial.ac.uk/media/imperial-college/medicine/sph/ide/gida-fellowships/Imperial-College-COVID19-NPI-modelling-16-03-2020.pdf The 'adaptive trigger' is discussed round about page 12.]
  13. No, it's from a paper by a group of researchers, mostly from Oxbridge and London universities, that was published in the BMJ (formerly British Medical Journal) a few years ago: https://bmjopen.bmj.com/content/7/11/e017722 I admit that I haven't read this paper yet, but I'll put it on my lockdown reading list.
  14. For most people, co-operation is always better than competition. At times like these, co-operation is the only option for everybody.
  15. I don't think so. You can't austerity yourself out of a depression. The 1930s proved that.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.