That Lancet Hydroxychloroquine Paper Withdrawal And Data Availability

13
1101

The Lancet has withdrawn the paper looking at hydroxychloroquine, the one that insisted that it increased, not reduced, the death rate. The problem is that no one can check the data. There’s suspicion – founded it appears – that some to much of that data was inaccurate, at best.

OK, this is good. Sure, it’s a pity that the trials were stopped and now it’s necessary to go back and do it all again. But hypothesis, then test it, if it fails against reality then drop the hypothesis. That’s science and while it does lead to error, as here, it’s the best way we’ve ever had of eventually zeroing in on what is actually true.

What interests though is this:

In a statement on Thursday, Mehra said: “Our independent peer reviewers informed us that Surgisphere would not transfer the full dataset, client contracts, and the full ISO audit report to their servers for analysis as such transfer would violate client agreements and confidentiality requirements. As such, our reviewers were not able to conduct an independent and private peer review and therefore notified us of their withdrawal from the peer-review process.”

That is, if the full data, the full model, ain’t available to be checked then the paper fails. Doesn’t matter about commercial confidentiality and all that, show your evidence or it ain’t science.

Which is a pretty good defence of that scientific method. The only thing is, well, have we seen this about Ferguson’s model of infection and lockdown? As far as I’m aware we’ve seen what he’s allowed us to see but not the specifics of the model itself, just some tangle of code that implements it.

And wasn’t there some climate change thing about emails and some university in East Anglia where access to all the underlying was stoutly denied?

That is, this new found insistence on doing science properly, as science. This is going to apply to everyone, right? Even Michael Mann?

13
Leave a Reply

avatar
8 Comment threads
5 Thread replies
13 Followers
 
Most reacted comment
Hottest comment thread
12 Comment authors
dodgy geezerjghPatSpikeJim Recent comment authors
  Subscribe  
newest oldest most voted
Notify of
Arthur Dent
Guest
Arthur Dent

Nope, Climate Science is different because, you know, “the science is settled”

Bloke in Germany
Guest
Bloke in Germany

The excuse is just handwavium from the Lancet who, despite being one of the most prestigious journals in the medical world, have form for publishing small, poorly-conducted garbage-in-garbage-out studies by publicity-hungry nonentities. There are plenty of situations where you cannot legally transfer databases to peer reviewers, client confidentiality and in this case patient confidentiality are reasonable explanations. All the anonymisation in the world isn’t enough to prevent patients being at risk of identification. This is not to defend Surgisphere, it’s just how things are. In the (industry-sponsored) clinical research world it’s not an issue, data that makes it into academic… Read more »

Leo Savantt
Guest
Leo Savantt

The Lancet’s problem, or at least one of its problems, is its editor in chief Richard Horton. A vocal Labour supporter and an anti-Trump left winger, who moans a lot about climate change, equality, the Palestinian cause and social justice etc. etc. etc. etc. He has a medical and psychology degree but his professorship is purely an honorary one. He has carried out no scientific research of note. It took him 12 years to retract Andrew Wakefield’s anti-vaccination paper which he never should have allowed to be published in the Lancet in the first place. He is completely unfit to… Read more »

John B
Guest
John B

Not disclosing data and model details is modus operandi for the prognosticators who predict the horrors of climate change, our diet, vehicle exhaust, etc.

They are charlatans, snake-oil purveyors but are well thought of by the Greta-Worship Cult – the political/intellectual/celebrity complex – and can do or say no wrong.

There is, I suspect, an element of Trump Derangement Syndrome at play too. He mentioned it as a possible remedy, therefore no effort must be spared to discredit him with all dispatch. Otherwise more care would have been taken over proper review of the paper before it was published.

Bloke in North Dorset
Guest
Bloke in North Dorset

This study was too good to check.

All their birthday’s came at once when they realised they had a chance to make Trump look stupid.

jgh
Guest
jgh

I was keeping a running tally of “death certificates issued stating Covid19 as cause of death” to back up my assertions that Belgium was the worst in the world, twice as bad as the UK. But, I’ve had people complaining that I should be tallying excess deaths compared with a 5-year average – which makes the UK the worst in the world, neck and neck with Spain. So I’m having to think about repeating three months of data gathering… but… ‘excess deaths’ is *all* deaths. So, falling down the stairs is a Covid19 death, drink-driving-homicide is a Covid19 death. I’m… Read more »

Quentin Vole
Guest
Quentin Vole

‘Excess deaths’ is the gold standard. Yes it will include falling down stairs, but there’s no reason for that to have increased over the 5-year average. Cause of death as written on the certificate can be notoriously inaccurate.

The UK is even quite quick (almost as fast as reasonably possible) in producing excess deaths data, but the trouble is that other countries are far more dilatory, making comparisons difficult until long after (a year, at least) the event.

Jim
Guest
Jim

Excess deaths over the average tells us how big a crisis we are going through, what it doesn’t tell us what is responsible for that crisis. A significant proportion of the excess deaths we are seeing are not the result of CV-19 directly, they are the result of our reaction to CV-19, the measures we have taken to attempt to deal with it, the lockdown etc, which is a completely different thing. Closing the NHS to virtually everyone bar CV cases for 2 months is certain to have killed thousands already, and will continue to do so for years.

Spike
Guest
Spike

Hear hear! Compare US statistics on the percentages of CV-19 deaths who were in long-term care facilities. (When CDC commandeered all the US’s health resources, nursing homes were a handy overflow tank for the infected, never mind that the captives would get sick and die, quarantined away from their loved ones.)

jgh
Guest
jgh

Ah ha! Today data has been released where the ‘excess deaths’ figures have been dug through and catagorised according to the actual cause of death. An initial glance is that about 80% of the 5-year-ED is from Covid19. I’ll have to stoke up some tea to plough the the figures.

dodgy geezer
Guest
dodgy geezer

Er – not exactly. I have been watching ‘excess deaths’ for several weeks now.

The point is that traffic accidents and falling downstairs are deaths which regularly ocurr – we can see their statistical averages. If there are ‘excess deaths’ that means something extra is happening. Something new.

This extra thing is almost certainly Covid-associated. Whether it’s a disease death or I e due to lockdown is, of course, another question.. .

Esteban
Guest
Esteban

To the extent that this is due to TDS (I suspect largely so) it really demonstrates how severe TDS is. President Trump mentioned HCQ in passing, just as a reassuring “there are a lot of possibilities for a cure or vaccine” statement. He didn’t bet the farm on it or praise it as THE ANSWER. If HCQ turns out to be extremely effective it wouldn’t have been a victory for President Trump, but the victims of TDS had to make this a hill to die over. Whom the Gods would destroy they first make mad.

Pat
Guest
Pat

Either evidence is presented to back an assertion or it is not. However valid the reason for non presentation of evidence, the assertion should be rejected without it, otherwise there is no reason to reject any assertion and we abandon the scientific method entirely.