|
After The Virus (pg. 6)
|
View this Thread in Original format
| SYSTEM-J |
| quote: | Originally posted by Paradox Lost
On an unrelated note, and not that this is in the grand scheme of things a major concern, but I don't know how the club scene rebounds from this any time soon. I can't imagine any promotion is gonna wanna book, say, John Digweed, fly him over here and pay his rate when he'll in all likelihood be playing to a half-empty dancefloor because people are still too afraid to go out, or at least cluster up. They can of course operate at dramatically reduced capacity and implement social distancing guidelines, but I don't think a smattering of people gyrating six feet apart makes for much of a vibe. Same with bars. You can drink much more affordably at home, but you go to a bar for the vibe, and what kind of vibe is an empty one? |
The only thing I'm worried about with clubs is when our government will allow them to re-open. A confined space where hundreds of people cram together for hours on end whilst categorically compromising their immune system through the triumvirate of no sleep, no food and lots of drugs seems like the worst possible place to allow open when you're battling a highly contagious virus.
Once they are allowed to re-open, I don't think there'll be any trouble filling them. Not here, anyway. The kind of people who are happy to buy bags of unknown and potentially lethal powder from petty criminals for a good time are not going to be deterred by Covid-19. |
|
|
| Boomer187 |
| quote: | Originally posted by SYSTEM-J
The only thing I'm worried about with clubs is when our government will allow them to re-open. A confined space where hundreds of people cram together for hours on end whilst categorically compromising their immune system through the triumvirate of no sleep, no food and lots of drugs seems like the worst possible place to allow open when you're battling a highly contagious virus.
Once they are allowed to re-open, I don't think there'll be any trouble filling them. Not here, anyway. The kind of people who are happy to buy bags of unknown and potentially lethal powder from petty criminals for a good time are not going to be deterred by Covid-19. |
Just take forehead temp before getting into club. Problem solved. :wtf: hehe |
|
|
| Dykes_on_Jay |
| You gotta pre-game on the risk taking behavior and moonshine pharmaceuticals in order to achieve maximum fist pump, bra. 65 pills beforehand means that hopefully the thermometer is rectal, because if simply standing makes you sweat, tattoo face will promptly, "nein. get out of line," you. Some payoff is better than none. |
|
|
| Arbiter |
It will be interesting to see how public trust in public health institutions such as the WHO and US CDC fares after the virus. Polls so far during the pandemic have shown pretty high levels of trust. That is a bit surprising to me, as their track record during the crisis seems to me to be very far from stellar.
For instance there is still no convincing explanation for the CDC's about-face on face masks. The CDC tried to explain its changed stance by citing "recent" studies about asymptomatic transmission, but that explanation is unconvincing since the studies it cited did little more than provide further evidence of what was already well known.
Perhaps an even more troubling example, the primary epidemiological model that the U.S. government has been relying on to guide its response to the pandemic, by the Institution for Health Metrics and Evaluation, is a hot mess, and that's putting it generously. After next-day deaths were shown to be outside the model's 95% confidence intervals 70% of the time, the only change they seem to have made is to broaden their own error range. The results now are so nonsensical that their current projections (which were updated today, by the way) predict a total of 60,709 U.S. fatalities COVID-19 by tomorrow. Considering that the U.S. recorded death count now stands at 61,656 as of this writing, the model's projection would require 947 dead COVID patients to come back to life tomorrow (in addition to no new deaths, of course). Suffice to say, I don't think they'll be quite right...
How does IHME's model reach such an unlikely conclusion? There are some clues in the preprint paper detailing its methodology. Their model is actually a very simplistic curve-fitting exercise. Effectively, they create a curve tracking the progression of the disease based on data from Wuhan (they've since integrated some data from Italy and the U.S. as well, but it still focuses primarily on Wuhan). They then assume that the decline in reported cases in Wuhan is exclusively a function of the number of days since government-imposed social distancing methods were put into place. From that they infer that enacting social distancing measures in other places (e.g., the United States or Europe) will cause the same decline in reported new cases and deaths as was observed in Wuhan. In short, they have demonstrated that if you apply very unrealistic assumptions to a small set of unreliable and incomplete data, you can bring people back to life. At least in a computer model. How supposed experts could still be advancing such an untenable methodology even after a month of being wrong time and time again should be a scandal in its own right. That they're being taken seriously by policymakers, doubly so.
And don't even get me started on the recent chatter about rushing vaccine candidates to human trials without proper safety protocols. That is bat insane on the level of causing a nuclear winter to ward off climate change.
In a rational population, you would think these and other blunders would lead to a crisis in confidence in our public health institutions. But so far there is no indication of such an effect. If anything, people have more confidence in the public health "science" being fed to them than ever. Will that continue even after the crisis passes? Or will people eventually come to view experts' response to the pandemic less charitably? And if they do eventually come to have a critical view of how the crisis was handled, will the response be to strengthen these institutions so they can (hopefully) do better next time, or to dismantle them (e.g., Trump's defunding of the WHO)? |
|
|
| Lews |
| quote: | Originally posted by Arbiter
It will be interesting to see how public trust in public health institutions such as the WHO and US CDC fares after the virus. Polls so far during the pandemic have shown pretty high levels of trust. That is a bit surprising to me, as their track record during the crisis seems to me to be very far from stellar.
For instance there is still no convincing explanation for the CDC's about-face on face masks. The CDC tried to explain its changed stance by citing "recent" studies about asymptomatic transmission, but that explanation is unconvincing since the studies it cited did little more than provide further evidence of what was already well known.
Perhaps an even more troubling example, the primary epidemiological model that the U.S. government has been relying on to guide its response to the pandemic, by the Institution for Health Metrics and Evaluation, is a hot mess, and that's putting it generously. After next-day deaths were shown to be outside the model's 95% confidence intervals 70% of the time, the only change they seem to have made is to broaden their own error range. The results now are so nonsensical that their current projections (which were updated today, by the way) predict a total of 60,709 U.S. fatalities COVID-19 by tomorrow. Considering that the U.S. recorded death count now stands at 61,656 as of this writing, the model's projection would require 947 dead COVID patients to come back to life tomorrow (in addition to no new deaths, of course). Suffice to say, I don't think they'll be quite right...
How does IHME's model reach such an unlikely conclusion? There are some clues in the preprint paper detailing its methodology. Their model is actually a very simplistic curve-fitting exercise. Effectively, they create a curve tracking the progression of the disease based on data from Wuhan (they've since integrated some data from Italy and the U.S. as well, but it still focuses primarily on Wuhan). They then assume that the decline in reported cases in Wuhan is exclusively a function of the number of days since government-imposed social distancing methods were put into place. From that they infer that enacting social distancing measures in other places (e.g., the United States or Europe) will cause the same decline in reported new cases and deaths as was observed in Wuhan. In short, they have demonstrated that if you apply very unrealistic assumptions to a small set of unreliable and incomplete data, you can bring people back to life. At least in a computer model. How supposed experts could still be advancing such an untenable methodology even after a month of being wrong time and time again should be a scandal in its own right. That they're being taken seriously by policymakers, doubly so.
And don't even get me started on the recent chatter about rushing vaccine candidates to human trials without proper safety protocols. That is bat insane on the level of causing a nuclear winter to ward off climate change.
In a rational population, you would think these and other blunders would lead to a crisis in confidence in our public health institutions. But so far there is no indication of such an effect. If anything, people have more confidence in the public health "science" being fed to them than ever. Will that continue even after the crisis passes? Or will people eventually come to view experts' response to the pandemic less charitably? And if they do eventually come to have a critical view of how the crisis was handled, will the response be to strengthen these institutions so they can (hopefully) do better next time, or to dismantle them (e.g., Trump's defunding of the WHO)? |
I haven't followed their model that closely (i.e., at all until now) but when I'm examining it currently it is predicting between 58,000-67,000 deaths in the US on 30 April. Data I'm seeing currently tells me there's been about 61,000 deaths so far. Seems like the model is relatively accurate? Where do you have historical data that 70% of the time deaths were outside their 95% confidence interval?
Frankly, I don't envy anyone trying to model this situation, or at least trying to model the effects of specific measures. But considering that the general public has no understanding of statistics or what a 95% confidence interval means, I'm not sure how much poor modelling will change people's perception of public health institutions. I'm simplistically going to assume that if deaths are not 'too bad,' then the public will think that the health authorities overreacted, while if deaths are atrocious the public will think the health authorities under-reacted. They're gonna be ed either way.
Really don't understand what your concern is with rushing vaccine candidates to human trials without 'proper safety protocols'. Chances are extremely high that the worst case scenario is that it kills the people given the vaccine candidates. |
|
|
| Arbiter |
The research paper about the IHME's high error rate is here:
https://arxiv.org/abs/2004.04734
I should note, however, that reporting on the paper actually slightly overstated its conclusions. While the IHME's model's performance was bad by any standard, it was only outside of 95% confidence intervals 70% of the time for certain subsets of the data (the overall range was that they were off 49-73% of the time). Their revised predictions avoid this type of error by providing very wide predictions for the possible number of next-day deaths. For example tomorrow's prediction for the entire U.S. will be within 95% confidence intervals as long as between -4,117 and 8,634 more people in the US die of COVID-19 today. That will no doubt be true, but it's not very informative.
Re "human challenge" vaccine trials, I don't have a problem with them in principle, but the goal of accelerating vaccine development is inherently problematic. Serious, even fatal side effects do not necessarily occur immediately after a vaccine is administered, so the length of the studies usually conducted both in animals and humans is a feature of vaccine development, not a bug. To their credit, Eyal et al. allude to this problem in passing, suggesting that participants "could continue to be followed longer term in parallel with the submission for licensure, so that suitable actions could be taken if any long-term adverse effects ... were identified." But they do not explain how this is an adequate substitute for the length of normal phase 3 trials. And since licensure is likely to be rushed as well, their suggestion seems to offer little additional protection before we begin mass-production and administration, at which point the existence of undetected long-term adverse effects would be a grievous error indeed.
I should also note that their assessment of the benefits of human challenge trials relies on the Imperial College model as supposedly supporting a possible death toll this year of 20 million (cited as ref 10 in the Harvard paper). This was not unreasonable when the article was submitted or even when published on March 31. But more recent data now strongly suggest that the rates of hospitalization and death are at least an order of magnitude lower than what the Imperial College assumed based on early and incomplete data from Wuhan and Italy. I wonder if they would have still advocated for human challenge trials with a more realistic estimate of how many lives a somewhat earlier vaccine might save. |
|
|
| Lews |
Is all the criticism about their 'next day' death count? I mean, just eyeballing it, it's volatile as all hell. I can't imagine how anyone could take that seriously in the first place. I was thinking that the criticism was about their overall forecast, which is seemingly more important/informative than their 'next day' forecast.
Full disclosure, I'm on the ethics board of a university that is running human vaccine trials and we aren't planning on recommending any successful vaccine for at least six months after seeing such success. Obviously long-term side effects could develop after that, it is impossible to know, but we're hoping to catch any short-to-medium term effects (presumably not all universities will be as careful as us, though). There simply isn't the time to run the normal extended tests we would like to see before recommending a drug. And, on the other hand, we also don't know what the long-term damage will be to people who catch and survive the virus (e.g., damaged lungs, etc). There's a lot of unknowns going on right now.
Where did the Imperial College model predict 20 million deaths? I saw 500,000 in the UK and 2.5 million in the US, or thereabouts, if no steps were taken. |
|
|
| Dykes_on_Jay |
| There is no way Arbiter's closet contains anything other than black slacks and turtlenecks. |
|
|
| Lews |
| quote: | Originally posted by Lews
Where did the Imperial College model predict 20 million deaths? I saw 500,000 in the UK and 2.5 million in the US, or thereabouts, if no steps were taken. |
Ahh, I was looking at the wrong Imperial College paper, my mistake.
Yeah, their death count (0.57%) seems a bit high to me, but not overly high. I've been thinking about 0.5% this whole time, though I obviously hope I'm mistaken. |
|
|
| planetaryplayer |
| NRG sword from covenant military |
|
|
|
|
