I find it puzzling how experts can disagree so profoundly about what seem to be relatively simple issues

[pmv]

Take this editorial from the British Medical Journal, no less

Covid-19: four fifths of cases are asymptomatic, China figures indicate

New evidence has emerged from China indicating that the large majority of coronavirus infections do not result in symptoms. Chinese authorities began publishing daily figures on 1 April on the number of new coronavirus cases that are asymptomatic, with the first day’s figures suggesting that...

www.bmj.com

In particular the comment from Tom Jefferson, an epidemiologist and honorary research fellow at the Centre for Evidence-Based Medicine at the University of Oxford that asks

If—and I stress, if—the results are representative, then we have to ask, ‘What the hell are we locking down for?’”

And contrast it with this reasoning by Dr William Hanage, a professor of the evolution and epidemiology of infectious disease at Harvard (so presumably every bit as expert as the first guy)

There have been more than 93,000 cases of Covid-19 identified in the UK. Let’s round that up and say it is 100,000. So if the reports from the BMJ editorial are accurate, the actual number would be that multiplied by five, in which case there would have already been half a million infections in the UK. If this really is the peak and we see as many cases on the way down as on the way up, that would total 1 million infections from the initial surge in the UK – hopefully all of those people would then be immune.

That would leave about 65 million people in the UK still without immunity.

I am going to be unusually optimistic here, and assume that everyone who has Covid-19 becomes fully immune (not a given), and that the virus is towards the less transmissible end of the range of estimates currently available. If this is the case, you would need half your population to have been infected to achieve a level of population immunity that would stop the epidemic continuing to grow and overwhelming healthcare systems.

As I write the UK is reporting more than 10,000 deaths from Covid-19. Due to the realities of collecting data during an infectious disease emergency like this, that is likely to be an underestimate. Again, if we assume this is the peak and there is the same number on the way down that’s 20,000 total from the initial surge. And to get to population immunity you have to multiply that by at least 30: based on the current data, that’s about 600,000 deaths to get there, minimum.

No matter how you crunch the numbers, this pandemic is only just getting started | William Hanage

People are understandably looking for good news. But the truth is, we’re nowhere near controlling coronavirus, says Harvard professor Dr William Hanage

www.theguardian.com

How can two qualified experts disagree so profoundly about something that doesn't on the face of it seem that complex an issue?

 

[pmv]

Though according to wiki, the first expert is, while clearly highly-qualified, regarded as an extreme outlier when it comes to the efficacy of the flu vaccine. Not sure if that's relevant.

In 2009 Jefferson was hired by the governments of Britain and Australia to update a systematic review of oseltamivir.[4] Initially he had a great deal of trouble getting results of the original trials.[4] Previous researchers had published trials without seeing the underlying data and it appeared that some of the published papers used ghost writers.[4] As the company, Roche, that did the trials refused to provide data for independent analysis, their conclusions in 2009 were that benefits could not be shown.[4]

In 2011 Jefferson's team was provided 22,000 pages of data from the European Medicines Agency and Roche eventually provided 3,000 pages of data.[4] In 2012, Jefferson was the lead author of another Cochrane review of this data which concluded that oseltamivir did not reduce the number of hospitalizations caused by influenza. Jefferson said that this review also found no evidence that oseltamivir stopped complications from the disease.[5]

His views about the effectiveness of influenza vaccines and his outspokenness about them are also controversial; at a 2007 meeting on pandemic preparedness he was shunned by other vaccine researchers and ate alone.[1] Reviews led by Jefferson have concluded that there is insufficient evidence to show that influenza vaccines reduce mortality or infection rates.[6] He has said that the studies claiming large reductions in mortality rates as a result of the influenza vaccine are "rubbish",[1] and that "influenza vaccines are about marketing and not science".[7] He has called repeatedly for placebo-controlled randomized clinical trials to prove the efficacy of flu vaccines, a position that most in the field hold as unethical.[1] His views on flu vaccines are opposed by the medical establishment.[1]

 

[sactoking]

It's quite simple: experts are still humans. Some experts have their views shaped by their expertise while others use their subject matter expertise to justify their views.

 

[Jaskalas]

How can two qualified experts disagree so profoundly about something that doesn't on the face of it seem that complex an issue?

Isn't it obvious? You are comparing apples to oranges.

The second guy did a standard by the numbers calculation. A pretty good one that I might stand by.

The first guy simply was not asked to do that, or does not care. Probably a fatalistic attitude, and why not? Clear evidence tells us that MORE than a majority of cases are asymptomatic. Meaning over a million MORE people than we have counted in the United States, think they are perfectly healthy, going to work, visiting with other people... and unwittingly spreading the virus.

The concept of containment is laughable given these figures. Yes, our lock-down can put an end to this peak. There will be another, and another. And another still. Until we achieve herd immunity. This WILL burn through our population and there is nothing we can do to stop it. How is that in contrast to the second "expert"? If anything his numbers reflect the cold hard truth where as the first guy parallels it by saying "!@#$ it".

Please help me identify their disagreement.

 

[pmv]

To be clear - the Harvard guy's explicit reasoning makes sense to me. In the absence of any counter-argument, I'd go with his take.

And without a lockdown, you will get those cases all coming at once, with the inevitable overwhelming of the medical system. With lock-downs you have a chance of achieving the curve-flattening thing, and also the chance of dragging out the process of the infection running through the population long enough that a vaccine will become available before you've had 600,000 dead. You get to herd-immunity via a vaccine, not via everyone eventually getting the disease.

But I'm just puzzled why the first guy doesn't seem to think through those numbers. He just suggests the lock-down is pointless. It wouldn't be a big deal if he were a crank or an expert in a completely different field, obviously suffering from delusions of expertise (as was the case with the Hoover Institute guys). But I find it a bit shocking that someone so well-qualified would be so cavalier, and still more so that he is uncritically-quoted being so in a prestigious journal like the BMJ. He either hasn't considered those numbers, or he doesn't care. Either way I'm a bit shocked.

It's all a bit disconcerting when people with real qualifications apparently say things that are wrong or at least grossly misleading (implying a lock-down serves no purpose).

 

[pmv]

Isn't it obvious? You are comparing apples to oranges.

The second guy did a standard by the numbers calculation. A pretty good one that I might stand by.

The first guy simply was not asked to do that, or does not care. Probably a fatalistic attitude, and why not? Clear evidence tells us that MORE than a majority of cases are asymptomatic. Meaning over a million MORE people than we have counted in the United States, think they are perfectly healthy, going to work, visiting with other people... and unwittingly spreading the virus.

The concept of containment is laughable given these figures. Yes, our lock-down can put an end to this peak. There will be another, and another. And another still. Until we achieve herd immunity. This WILL burn through our population and there is nothing we can do to stop it. How is that in contrast to the second "expert"? If anything his numbers reflect the cold hard truth where as the first guy parallels it by saying "!@#$ it".

Please help me identify their disagreement.

But that ignores the likelihood of a vaccine being developed. Vaccines are how you are supposed to reach herd-immunity, not via everyone becoming infected and recovering or dying. Aside from that the 'curve flattening' effect of lock-downs still applies.

 

[pmv]

It's quite simple: experts are still humans. Some experts have their views shaped by their expertise while others use their subject matter expertise to justify their views.

Well that may be the answer, though I would have hoped that at least in some disciplines, and at least with people really at the top of their fields, the latter is very rare. Otherwise the whole concept of expertise becomes rather shaky.

 

[mikeymikec]

I'm kinda loathe to ask this question about an expert, but is he fucking dense or what?

- edit - pmv already has made my point while I was drafting my response

 

[JEDIYoda]

It's quite simple: experts are still humans. Some experts have their views shaped by their expertise while others use their subject matter expertise to justify their views.

you were nicer than I would have been....
My take is that is the difference when the Democrats and the republican view the same event and come away with two total different viewpoints...lolol

 

[Grooveriding]

I work around people doing strictly research. The best of them almost universally are able to put aside their biases and adhere to forced objectivity, this is not true of all of them! Also, there is nothing in either link that is a disagreement on any facts that I could see, but there is a clear difference of opinion on approach.

Many people are asymptomatic with this virus, this is even more of a reason for continued social distancing, because testing everyone is just not feasible at this time. So these asymptomatic carriers are highly dangerous to those of us who are susceptible to being symptomatic post-infection. They don't know they're carrying it, no one else can tell they may be sick from their visible condition; so they can be out and about spreading the virus like mad.

This virus is more contagious and more lethal than the standard flu. It needs to be contained or it will kill millions. We'll get some good data on this if Trump manages to convince everyone to end the US lockdown in a few weeks. There will be another significant jump in infections and deaths most likely. Comparatively, other countries are planning to continue past this date.

 

[HomerJS]

Seems to be another reason for widespread and regular testing. Symptomatic at least the person and others know to avoid contact. If asymptomatic people are regularly tested distancing can be maintained for the people who need it until a vaccine is developed.

 

[abj13]

I think the most important question is how do you define an expert? Tom Jefferson is an expert in metaanalysis of existing studies, with a particular focus on vaccines and antivirals. In one sense that makes him an expert in biology, infectious diseases, viral infection, vaccination, treatment etc. On the other hand, does that make him an expert in viral transmission, outbreak containment, and mitigation? Not really. Is he really the best expert to discuss the effectiveness of social distancing and making a decision when to end it? I really don't see a publication that would justify that he's the best person to speak on that topic.

Jefferson is also a very bombastic, in-your-face type of scientist. He's more willing to write editorials and tell other scientists how they are wrong in a not so gentle way. Of course, science needs to be scrutinized and heavily reviewed and retested. But often, Jefferson will ruffle some feathers in the ways he goes about it.

Then of course, you have to think about the content of his statements and whether it holds up to any scrutiny. I doubt the BMJ asked him the obvious question in response to his statement: "Jefferson said that it was quite likely that the virus had been circulating for longer than generally believed and that large swathes of the population had already been exposed." If what Jefferson says is true, wouldn't that mean the there should have been already a huge peak in hospitalizations and deaths? The timeline doesn't make any logical sense. The virus doesn't magically only cause asymptomatic infections in the beginning, with very few hospitalizations/death. There's no preexisting immunity, so those who are most likely to have complications are probably just as likely to be infected with COVID-19 as those who would develop asymptomatic infection in the beginning of the outbreak.

I would guess that if someone were to ask him those type of questions, I'm would predict he'd walk back his statements about "locking down."

 

[Moonbeam]

"Jefferson said that it was quite likely that the virus had been circulating for longer than generally believed and that large swathes of the population had already been exposed."

Can you show where he makes that claim? I didn’t see it. If he did your counterargument wins in my opinion.

 

[abj13]

Can you show where he makes that claim? I didn’t see it. If he did your counterargument wins in my opinion.

That line is directly from the article. Now, maybe the BMJ misquoted him, but this is how the article was written:

Covid-19: four fifths of cases are asymptomatic, China figures indicate

New evidence has emerged from China indicating that the large majority of coronavirus infections do not result in symptoms. Chinese authorities began publishing daily figures on 1 April on the number of new coronavirus cases that are asymptomatic, with the first day’s figures suggesting that...

www.bmj.com

Tom Jefferson, an epidemiologist and honorary research fellow at the Centre for Evidence-Based Medicine at the University of Oxford, said the findings were “very, very important.” He told The BMJ, “The sample is small, and more data will become available. Also, it’s not clear exactly how these cases were identified. But let’s just say they are generalisable. And even if they are 10% out, then this suggests the virus is everywhere. If—and I stress, if—the results are representative, then we have to ask, ‘What the hell are we locking down for?’”

Jefferson said that it was quite likely that the virus had been circulating for longer than generally believed and that large swathes of the population had already been exposed.

 

[Moonbeam]

That line is directly from the article. Now, maybe the BMJ misquoted him, but this is how the article was written:

Covid-19: four fifths of cases are asymptomatic, China figures indicate

New evidence has emerged from China indicating that the large majority of coronavirus infections do not result in symptoms. Chinese authorities began publishing daily figures on 1 April on the number of new coronavirus cases that are asymptomatic, with the first day’s figures suggesting that...

www.bmj.com

Tom Jefferson, an epidemiologist and honorary research fellow at the Centre for Evidence-Based Medicine at the University of Oxford, said the findings were “very, very important.” He told The BMJ, “The sample is small, and more data will become available. Also, it’s not clear exactly how these cases were identified. But let’s just say they are generalisable. And even if they are 10% out, then this suggests the virus is everywhere. If—and I stress, if—the results are representative, then we have to ask, ‘What the hell are we locking down for?’”

Jefferson said that it was quite likely that the virus had been circulating for longer than generally believed and that large swathes of the population had already been exposed.

Thanks. I did read that I now realize without it setting off an alarm. It is the same supposition from the Hoover Institute guy and has been debunked according to my understanding. The virus has been tracked back to China to a single animal to human jump by analysis of mutated strains. There was no earlier wide spread infection according to that data.

We may just be looking at another person more worried about the economic effects of social isolation than the lives it may save and the chaos prevented by curve flattening. Hard to say, but seems his science is wrong.

 

[shortylickens]

Science is not an absolute.
Humans are imperfect. They learn, then they learn some more.
We had all kinds of notions about medicine in 1950 that are now considered out of date and sometimes just plain wrong. Time marches on. People do more experiments. We know more than we used to.
As far as two scientists believing completely different things at the exact same time, look at their testing methods and ask around in their community. I bet they are doing things differently.

 

[Jaskalas]

But that ignores the likelihood of a vaccine being developed. Vaccines are how you are supposed to reach herd-immunity, not via everyone becoming infected and recovering or dying. Aside from that the 'curve flattening' effect of lock-downs still applies.

Ah, you are correct regarding flattening the curve. The first guy is ignoring the benefit of that. And I overlooked it in my observation of their comments.

As for immunity. Both recovery and vaccine achieve it, but the vaccine would exist year(s) later. Not when it is needed most.

 

[ondma]

I just read the entire article. Seems pretty "unscientific" for what I thought was a respected medical journal. The article is making a very large extrapolation from a small set of data.
First, they say 130 of 166 infections were asymptomatic. OK, but how many total people did they test to find the 166 total infections? Simply because 130 of 166 were asymptomatic do not in any way indicate that the virus is widespread in the entire population. That can only be determined by testing a representative sample of the entire population and seeing how many have active infection (with or without symptoms) or have been exposed and recovered.

 

[pmv]

Ah, you are correct regarding flattening the curve. The first guy is ignoring the benefit of that. And I overlooked it in my observation of their comments.

As for immunity. Both recovery and vaccine achieve it, but the vaccine would exist year(s) later. Not when it is needed most.

I don't believe, though, that any expert thinks its a good idea to achieve 'herd immunity' by letting everyone be infected. That's just not the normal view. Vaccines are how its normally done. Also I don't know that the time scale is going to be that different, unless you abandon all efforts at curve-flattening, in which case there will be huge numbers of deaths.

abj13 makes a good point that Jefferson's field seems to not directly be epidemiology. Also that BMJ article is an editorial not a peer-reviewed paper. Still, though, I'm taken aback that they aren't both more careful in what they say in a situation like this, with the status they have. It really confuses laypeople like me.

(For a moment I was thinking that the BMJ published Andrew Wakefield's infamous flawed vaccine paper, but, seems not, that was The Lancet, the other top UK medical journal)

I wonder if there's a connection with Jefferson's existing animus towards flu vaccines? It doesn't appear that he's in any way in the Wakefield camp regarding vaccines in general, in fact, for all I know he's the "good guy" in that particular argument - wiki seemed to imply the vaccine manufacturers had been less-than-forthcoming with the data. But sounds like his stance is a bit controversial. Sheesh, I thought once things were established by published 'studies' everyone got on board with the concensus, it doesn't sound as if it works like that in that field.

Made me laugh that at conferences his fellow researchers refuse to talk to him and he 'has to eat alone' (according to wiki). I was wondering if they also steal his dinner money and give him wedgies?

 

[Moonbeam]

Ah, you are correct regarding flattening the curve. The first guy is ignoring the benefit of that. And I overlooked it in my observation of their comments.

As for immunity. Both recovery and vaccine achieve it, but the vaccine would exist year(s) later. Not when it is needed most.

Timeline is not known for a vaccine. One may be as chose as 8 months. An anti-viral is also showing signs it can treat bad cases.

 

[Jaskalas]

Timeline is not known for a vaccine. One may be as chili’s as 8 months.

Yeah, next year at the earliest.

Given how contagious this thing is... wouldn't it... cover more than half our population by then?
Lockdown is the only thing slowing it down, but we can't shut everyone down all year.

 

[Commodus]

This is a reminder that it's not enough to conduct a scientific test; it also has to be founded on solid science that collects adequate data and avoids making assumptions. And presuming much higher asymptomatic rates seems flawed when we have evidence of high symptomatic rates in places like that Smithville pork plant.

 

[Moonbeam]

Yeah, next year at the earliest.

Given how contagious this thing is... wouldn't it... cover more than half our population by then?
Lockdown is the only thing slowing it down, but we can't shut everyone down all year.

Next year as far as we know now and according to the current state of our thinking.and according to the level of interest in a cure the President decides he knows best. We can't have doctors and scientists and research universities running things. Billions have been spent denying scientific reasoning. It just fucks up authoritarianism and the breainwashing of the population. Can't have trained thinkers doing our shopping. How many pink flamingos will you sell that way. You seem intent on saving a system that should never have existed but for self hate and it's companion, human greed.

 

[Jaskalas]

This is a reminder that it's not enough to conduct a scientific test; it also has to be founded on solid science that collects adequate data and avoids making assumptions. And presuming much higher asymptomatic rates seems flawed when we have evidence of high symptomatic rates in places like that Smithville pork plant.

Please reference where you learned that the entirety of the Smithfield workers were tested. You are the first I have heard of that, and no article I read seems to indicate that "fact". What I do have are 3 different sources indicating that being asymptomatic happens in a MAJORITY of COVID-19 cases. This topic presented the third source.

• Iceland random sampling of population.
• U.S. Aircraft Carrier entire crew testing.
• Four fifths of cases are asymptomatic, China figures indicate

I would find it fascinating if the pork plant workers are sicker than the average population. Being asymptomatic may be tied into age or other demographic factors.

 

[pmv]

One other point I don't understand - is it known that asymptomatic cases then gain immunity when they recover?

The research from China (which may-or-may-not be reliable, given that both the university research team and the newspaper it was reported in, are Chinese, and hence I would assume are subject to the political influences that one suspects affects everything in China) seemed to say that many mild cases failed to develop much of an antibody response.

If the large number of asymptomatic cases in fact don't gain much of an immune benefit from their infection, would that not essentially erase the significance of there being a lot of asymptomatic cases in the first place? (Other than the bad side of it, which is it makes it harder to stop the spread).