The pseudoscience behind the politicized strategies to mitigate this pandemic remind me of the pseudoscience of economic theory. When the real world doesn't conform to the textbooks, do we still cling to the mathematics that support neoclassical (or neoliberal) economic theory?
I was struck by another Washington Post article headline yesterday: "Booster shots won’t stop the delta variant. Here’s the math to prove it." Their tagline was this: "Vaccinating more people would help more than giving another dose to the vaccinated"
Delving in to R0 and Re mathematics, they explained "The Re helps us estimate how much a disease may spread when a population has at least some immunity."
Never mind that "some immunity" includes all those people who are vaccinated that are increasingly getting breakthrough infections.
The Lancet has had its share of controversies during this pandemic. But this article is a good one to read (except we can all cough when reading "relatively low mutation rate of SARS-CoV-2"): SARS-CoV-2 evolution and vaccines: cause for concern?
"From this overview, only data on influenza might suggest that evolution in SARS-CoV-2 could eventually lead to a less efficacious vaccine. A protective factor is the relatively low mutation rate of SARS-CoV-2, although prolonged infection in immunocompromised hosts might accelerate mutation.5 However, the length of the spike protein used by licensed vaccines is relatively short (∼1270 amino acids), and one preprint paper has indicated that the natural antibody response to infection (and presumably also to a spike protein-based vaccine) is concentrated in just two sections of the protein: the N-terminal domain (NTD) and receptor-binding domain (RBD).6... Another notable finding is that SARS-CoV-2 passaged in the presence of polyclonal antibodies (in the form of convalescent sera) can also mutate and escape neutralisation by the multiple antibodies. In a series of experiments described in a 2020 preprint, SARS-CoV-2 was grown in the presence of neutralising COVID-19 convalescent plasma from a recovered patient.9 After serial passages, three mutations were generated (a deletion and insertion in the NTD loops, and a point substitution in the RBD) that allowed the newly formed variant to become completely resistant to plasma neutralisation... An unresolved question is the effect that the reported mutations might have on T cell immunity, which is known from vaccine trials to be robustly stimulated by the recombinant spike protein. T cell escape has been well described for persisting viruses such as HIV, HBV, and hepatitis C virus.15"
"What’s the mechanism behind why some people are developing these syndromes? Chris Smith spoke to Yale Immunologist Akiko Iwasaki...
Akiko - So there are a couple of theories that are now posed to explain long COVID. One is a lingering virus or a viral reservoir that persists in a person that can stimulate chronic inflammation. The other possibility is autoimmunity; that even a mild viral infection can trigger autoimmunity, which has long-term consequences."
Both of these hypotheses are concerning. But when the virus lingers and/or evade detection (think immunocompromised hosts in a hospital setting; in a nursing home or in high-density living), we are in big trouble when public health agencies assert that full vaccination is the key to ending this pandemic, and tell us to doff masks and put full faith in rapid tests and vaccine status.
I do think we have to prepare for the failures we had in developing HIV vaccines to control this pandemic, with attention to pre-existing candidate antiviral drugs and existing ones, like ivermectin, unless it too becomes resistant to mutant strains.
I agree that R0 is way over interpreted since, as you say, it is a major oversimplification that does not relate to the actual rate of transmission that of course depends on many factors that change over time and place.
Many epidemiologists seem to like it as a general measure that can give a sense of relative transmissibility between different viruses, and it makes life easy in SIR-based simulators.
But modeling based on R0 is hopelessly simplistic, which I suppose is one of your key points (besides your argument that it is used to hype up the danger of the pandemic).
I think that estimating and tracking Rt over time and place, and assessing which factors may be impacting it, can have some value if it is done well, but of course people doing such modeling need to not take themselves too seriously, understanding all of the uncertainty and various factors impacting this. So I am not sure if you are saying that any assessment of Rt over time is meaningless, but if you are I'd disagree.
Also, Eric's tweets were a bit over the top, but was he really wrong? In January almost no one anticipated this local outbreak in Wuhan would spread all of the world and become so hard to contain, leading to countries all over the world instituting lockdowns and other measures we've never seen, motivating rapid development and dissemination of vaccines all over the world, and dominating the world for the better part of two years so far.
No matter what you think about the vaccines, or the mitigation strategies that have been employed, or how the media has portrayed the crisis, do you really deny that SARS-CoV-2 transmits remarkably efficiently, plus spreads during the pre-symptomatic period, making it hard to contain, right?
And with all the talk of immune escape variants, it is the more transmissible variants, first D614G, then Alpha, and now Delta, that have out-competed the others and became the dominant variants -- and these seem to be transmitting even more rapidly than the original Wuhan strain (and there are plausible biological explanations for how the mutations that characterize them, especially in the RBD, would lead to greater transmission. So far, these variants, which have mild immune escape properties (as measured by Nabs) have out-competed the variants with the strongest immune escape properties (e.g. Beta and Gamma). This may change especially as a higher proportion of the world is either vaccinated or previously infected with previous strains, so maybe immune escape will become the determining factor for natural selection, but right now it still appears transmissibility is the more major factor.
But even if delivered in a hysterical manner and quoting estimates of transmissibility that may have been inflated (and certainly by their nature oversimplified), an early prediction that SARS-CoV-2 would become hard to contain sounds pretty prescient to me.
As usual, you are wrong on most all accounts, including whether I think R can be made useful (you didn't read well enough). But some of that will take additional articles.
You have unanswered questions and discussions in your email that is weeks old that explains numerous reasons why the vaccine campaigns are hurting. You just choose to ignore them. So be it.
When I wrote you several thousand words in personal email, and offered to take my personal time with more, you told me you didn't need any of that...because you are a Professor of Biostatistics.
None of this can be explained succinctly or the entire article series would have been complete months ago.
And it doesn't matter what I write, because to you it's not an argument until it's peer reviewed (unless you make it), and the official narrative is correct (even if it doesn't square with the high correlation between vaccinations and death internationally, or totally lack of correlation in cases in U.S. counties...or the CDC admitting that it fabricated numbers, and so on and so on).
You're just going to have to be patient now, and that's you're own fault.
But again, as I've recommended before, go back and read the first email I sent you. If you can manage not to black it all out, there are a lot of answers in there for you.
Better...how about you take your own blog posts...and publicly admit that you haven't bothered to examine the possibility of survivor bias in the numbers regarding deaths.
Yes, I know you were offended that I responded abruptly and cut off our discussion that day and have never gotten over that. I tried to apologize and explain was going on with me that day. But if you want to use that as a reason to not respond to my questions, OK that is your prerogative.
But if you wanted to, you could lay out summary bullet points outlining the key aspects of your argument. Nothing is so complex that you can't at least list key bullet points or summary points or key ideas behind the argument. Complexity is not an excuse for lack of brevity -- in general when really understands their points they should be able to state them succinctly.
I've not said anything to you about the need for peer review, and I am not sure why you feel the need to tell me what my response will be to your argument before you even make it. And in my writing I try to transparently lay out my arguments and give citations where possible, and acknowledge uncertainty where possible.
And I'd be glad to engage in a discussion about survivor bias -- I have mentioned this as potential issue. Your argument is that the apparent vaccine effectiveness vs. hospitalizations/severe disease is an artifact of survivor bias whereby the vaccines "kill off" the weak older people immediately after vaccination, so they don't appear in the data set, and then the "weak" among the vaccinated are out of the sample when it comes to evaluating future severe disease, but the unvaccinated still have these "weak" and so they look worse. That's the gist right?
I am fine with that as a hypothesis, and I've seen your arguments suggesting many classified COVID deaths after vaccination are in fact caused by vaccine and misclassified. I am open to that possibility, but am not convinced. That is a very strong assertion, not just showing it happened but also showing the magnitude is enough to skew the vaccine effectiveness estimates for severe disease, which in my opinion you have not sufficiently demonstrated for me to be convinced.
I am open to more data and arguments on the factor and always willing to assess the evidence you provide. I can't promise I will be convinced but always glad to listen and consider.
"But if you want to use that as a reason to not respond to my questions, OK that is your prerogative."
Again, you're making up a narrative instead of reading my words. What a waste of time.
"But if you wanted to, you could lay out summary bullet points outlining the key aspects of your argument. Nothing is so complex that you can't at least list key bullet points or summary points or key ideas behind the argument."
False. And it's worse in the case of talking to a narrative gatekeeper. For instance, you've often started with false claims like "the variants appeared before vaccination", and the resulting conversation clear that up takes three times as long as the article.
"I am fine with that as a hypothesis, and I've seen your arguments suggesting many classified COVID deaths after vaccination are in fact caused by vaccine and misclassified. I am open to that possibility, but am not convinced."
If your first reaction is to "be unconvinced" rather than recognize that the very first responsibility of everyone involved (including us as statistical watchdogs who should *not* EVER write about statistics without clear indication of known unknowns) is to lay out a system that ensures that study and documentation takes place, then you allow yourself to be a tool for institutional capture. First principle come first, right? If the reason we don't have good data is that the virtopsy operators were told to stand down, and the neurologists seeing 20x patients haven't had time to enter more than 2 or 2,000 cases into VAERS, then the system is broken and the statistics aren't real.
If you start anywhere else, you throw the precautionary principle out the window along with government and medical transparency. That's a recipe for disaster whether it happens here and now, or somewhere down the line.
This is the totalitarian line. You're not a statistician if you don't defend the need to connect numbers to reality.
The pseudoscience behind the politicized strategies to mitigate this pandemic remind me of the pseudoscience of economic theory. When the real world doesn't conform to the textbooks, do we still cling to the mathematics that support neoclassical (or neoliberal) economic theory?
https://www.washingtonpost.com/news/storyline/wp/2015/01/05/the-protesters-who-are-trying-to-upend-the-fantasy-world-of-economics/
I was struck by another Washington Post article headline yesterday: "Booster shots won’t stop the delta variant. Here’s the math to prove it." Their tagline was this: "Vaccinating more people would help more than giving another dose to the vaccinated"
https://www.washingtonpost.com/outlook/coronavirus-vaccine-booster-shots/2021/08/11/aefec5dc-fae0-11eb-9c0e-97e29906a970_story.html
Delving in to R0 and Re mathematics, they explained "The Re helps us estimate how much a disease may spread when a population has at least some immunity."
Never mind that "some immunity" includes all those people who are vaccinated that are increasingly getting breakthrough infections.
The Lancet has had its share of controversies during this pandemic. But this article is a good one to read (except we can all cough when reading "relatively low mutation rate of SARS-CoV-2"): SARS-CoV-2 evolution and vaccines: cause for concern?
https://www.thelancet.com/journals/lanres/article/PIIS2213-2600(21)00075-8/fulltext
"From this overview, only data on influenza might suggest that evolution in SARS-CoV-2 could eventually lead to a less efficacious vaccine. A protective factor is the relatively low mutation rate of SARS-CoV-2, although prolonged infection in immunocompromised hosts might accelerate mutation.5 However, the length of the spike protein used by licensed vaccines is relatively short (∼1270 amino acids), and one preprint paper has indicated that the natural antibody response to infection (and presumably also to a spike protein-based vaccine) is concentrated in just two sections of the protein: the N-terminal domain (NTD) and receptor-binding domain (RBD).6... Another notable finding is that SARS-CoV-2 passaged in the presence of polyclonal antibodies (in the form of convalescent sera) can also mutate and escape neutralisation by the multiple antibodies. In a series of experiments described in a 2020 preprint, SARS-CoV-2 was grown in the presence of neutralising COVID-19 convalescent plasma from a recovered patient.9 After serial passages, three mutations were generated (a deletion and insertion in the NTD loops, and a point substitution in the RBD) that allowed the newly formed variant to become completely resistant to plasma neutralisation... An unresolved question is the effect that the reported mutations might have on T cell immunity, which is known from vaccine trials to be robustly stimulated by the recombinant spike protein. T cell escape has been well described for persisting viruses such as HIV, HBV, and hepatitis C virus.15"
This potential to evade vaccines includes a potential to evade detection. https://www.nature.com/articles/s41587-021-00845-3
With a pivot to "What's causing long COVID?"
https://www.thenakedscientists.com/articles/interviews/whats-causing-long-covid
"What’s the mechanism behind why some people are developing these syndromes? Chris Smith spoke to Yale Immunologist Akiko Iwasaki...
Akiko - So there are a couple of theories that are now posed to explain long COVID. One is a lingering virus or a viral reservoir that persists in a person that can stimulate chronic inflammation. The other possibility is autoimmunity; that even a mild viral infection can trigger autoimmunity, which has long-term consequences."
Both of these hypotheses are concerning. But when the virus lingers and/or evade detection (think immunocompromised hosts in a hospital setting; in a nursing home or in high-density living), we are in big trouble when public health agencies assert that full vaccination is the key to ending this pandemic, and tell us to doff masks and put full faith in rapid tests and vaccine status.
I do think we have to prepare for the failures we had in developing HIV vaccines to control this pandemic, with attention to pre-existing candidate antiviral drugs and existing ones, like ivermectin, unless it too becomes resistant to mutant strains.
I appreciate your work sir. How you manage to do it all baffles me.
Lies, damned lies, and statistics.
I agree that R0 is way over interpreted since, as you say, it is a major oversimplification that does not relate to the actual rate of transmission that of course depends on many factors that change over time and place.
Many epidemiologists seem to like it as a general measure that can give a sense of relative transmissibility between different viruses, and it makes life easy in SIR-based simulators.
But modeling based on R0 is hopelessly simplistic, which I suppose is one of your key points (besides your argument that it is used to hype up the danger of the pandemic).
I think that estimating and tracking Rt over time and place, and assessing which factors may be impacting it, can have some value if it is done well, but of course people doing such modeling need to not take themselves too seriously, understanding all of the uncertainty and various factors impacting this. So I am not sure if you are saying that any assessment of Rt over time is meaningless, but if you are I'd disagree.
Also, Eric's tweets were a bit over the top, but was he really wrong? In January almost no one anticipated this local outbreak in Wuhan would spread all of the world and become so hard to contain, leading to countries all over the world instituting lockdowns and other measures we've never seen, motivating rapid development and dissemination of vaccines all over the world, and dominating the world for the better part of two years so far.
No matter what you think about the vaccines, or the mitigation strategies that have been employed, or how the media has portrayed the crisis, do you really deny that SARS-CoV-2 transmits remarkably efficiently, plus spreads during the pre-symptomatic period, making it hard to contain, right?
And with all the talk of immune escape variants, it is the more transmissible variants, first D614G, then Alpha, and now Delta, that have out-competed the others and became the dominant variants -- and these seem to be transmitting even more rapidly than the original Wuhan strain (and there are plausible biological explanations for how the mutations that characterize them, especially in the RBD, would lead to greater transmission. So far, these variants, which have mild immune escape properties (as measured by Nabs) have out-competed the variants with the strongest immune escape properties (e.g. Beta and Gamma). This may change especially as a higher proportion of the world is either vaccinated or previously infected with previous strains, so maybe immune escape will become the determining factor for natural selection, but right now it still appears transmissibility is the more major factor.
But even if delivered in a hysterical manner and quoting estimates of transmissibility that may have been inflated (and certainly by their nature oversimplified), an early prediction that SARS-CoV-2 would become hard to contain sounds pretty prescient to me.
As usual, you are wrong on most all accounts, including whether I think R can be made useful (you didn't read well enough). But some of that will take additional articles.
You have unanswered questions and discussions in your email that is weeks old that explains numerous reasons why the vaccine campaigns are hurting. You just choose to ignore them. So be it.
I guess I shouldn't be surprised you are not willing to make succinct comments on what you disagree with and why. Oh well.
When I wrote you several thousand words in personal email, and offered to take my personal time with more, you told me you didn't need any of that...because you are a Professor of Biostatistics.
None of this can be explained succinctly or the entire article series would have been complete months ago.
And it doesn't matter what I write, because to you it's not an argument until it's peer reviewed (unless you make it), and the official narrative is correct (even if it doesn't square with the high correlation between vaccinations and death internationally, or totally lack of correlation in cases in U.S. counties...or the CDC admitting that it fabricated numbers, and so on and so on).
You're just going to have to be patient now, and that's you're own fault.
But again, as I've recommended before, go back and read the first email I sent you. If you can manage not to black it all out, there are a lot of answers in there for you.
Better...how about you take your own blog posts...and publicly admit that you haven't bothered to examine the possibility of survivor bias in the numbers regarding deaths.
Yes, I know you were offended that I responded abruptly and cut off our discussion that day and have never gotten over that. I tried to apologize and explain was going on with me that day. But if you want to use that as a reason to not respond to my questions, OK that is your prerogative.
But if you wanted to, you could lay out summary bullet points outlining the key aspects of your argument. Nothing is so complex that you can't at least list key bullet points or summary points or key ideas behind the argument. Complexity is not an excuse for lack of brevity -- in general when really understands their points they should be able to state them succinctly.
I've not said anything to you about the need for peer review, and I am not sure why you feel the need to tell me what my response will be to your argument before you even make it. And in my writing I try to transparently lay out my arguments and give citations where possible, and acknowledge uncertainty where possible.
And I'd be glad to engage in a discussion about survivor bias -- I have mentioned this as potential issue. Your argument is that the apparent vaccine effectiveness vs. hospitalizations/severe disease is an artifact of survivor bias whereby the vaccines "kill off" the weak older people immediately after vaccination, so they don't appear in the data set, and then the "weak" among the vaccinated are out of the sample when it comes to evaluating future severe disease, but the unvaccinated still have these "weak" and so they look worse. That's the gist right?
I am fine with that as a hypothesis, and I've seen your arguments suggesting many classified COVID deaths after vaccination are in fact caused by vaccine and misclassified. I am open to that possibility, but am not convinced. That is a very strong assertion, not just showing it happened but also showing the magnitude is enough to skew the vaccine effectiveness estimates for severe disease, which in my opinion you have not sufficiently demonstrated for me to be convinced.
I am open to more data and arguments on the factor and always willing to assess the evidence you provide. I can't promise I will be convinced but always glad to listen and consider.
"But if you want to use that as a reason to not respond to my questions, OK that is your prerogative."
Again, you're making up a narrative instead of reading my words. What a waste of time.
"But if you wanted to, you could lay out summary bullet points outlining the key aspects of your argument. Nothing is so complex that you can't at least list key bullet points or summary points or key ideas behind the argument."
False. And it's worse in the case of talking to a narrative gatekeeper. For instance, you've often started with false claims like "the variants appeared before vaccination", and the resulting conversation clear that up takes three times as long as the article.
"I am fine with that as a hypothesis, and I've seen your arguments suggesting many classified COVID deaths after vaccination are in fact caused by vaccine and misclassified. I am open to that possibility, but am not convinced."
If your first reaction is to "be unconvinced" rather than recognize that the very first responsibility of everyone involved (including us as statistical watchdogs who should *not* EVER write about statistics without clear indication of known unknowns) is to lay out a system that ensures that study and documentation takes place, then you allow yourself to be a tool for institutional capture. First principle come first, right? If the reason we don't have good data is that the virtopsy operators were told to stand down, and the neurologists seeing 20x patients haven't had time to enter more than 2 or 2,000 cases into VAERS, then the system is broken and the statistics aren't real.
If you start anywhere else, you throw the precautionary principle out the window along with government and medical transparency. That's a recipe for disaster whether it happens here and now, or somewhere down the line.
This is the totalitarian line. You're not a statistician if you don't defend the need to connect numbers to reality.