First, the best thing an adult can do to lower their own risk of bad covid outcomes is get vaccinated
Second, over the next decade it is inevitable that all people on earth will encounter the virus. Just a matter of time, but cannot be avoided
Third, after vaccination, you will have as good chances as you can possibly have
Fourth, life is not zero risk; never has been, never will be
We have always accepted risk, and we will yet again (many already are)
Fifth, the more you postpone or delay things that make us human (after vaccination) the more you suffer, spiritually and mentally, and may not even recognize it
Sixth, cloth masks did not work in an unvaccinated population in bangladesh, only surgical masks worked
Seventh, this RCT does not generalize to vaccinated cohorts, but if you wish to generalize, I would not cloth mask
Eight, professors reluctant to teach b/c vaccinated students unmasked should consider that they cannot control the environment in all aspects of life...
They will get exposed.
And since cloth masking failed, whether or not kids cloth mask will not even change time to exposure.
Ninth, difficulty in resuming life is why nearly all pandemic guidance prior to 2015 warned to engage in the least disruptions possible...
..
...
..
whoops
Tenth, testing asymptomatic, vaccinated people and finding sars-cov-2 is unlikely to be useful in most settings, and certainly not young people who are vaccinated
These companies should be running RCTs, instead they got market share easily
Eleventh, the more you disrupt life for PCR or antigen positive results without significant numbers of ill people, the more you pay the price of restrictions without possibility of countervailing benefit
Twelfth, you can screw up many parts of life, but screwing up school for young kids any longer is untenable
Thirteenth, if you distort risk perception, then society will be unable to think clearly about myocarditis, boosters and vaccinating kids...
...
....
...
whoops again
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Every single one of these garbage papers use the denominator of PCR+ covid infections, and not the actual denominator of people who had covid-- many do not present to the doc. Every single one of these papers is shit. I would be embarrassed to be an author. Incompetent work.
People with covid who are so sick they have to go to the doctor have poorer health than people who don't. That should be in the journal of obvious things. All of these authors are extrapolating beyond the evidence. They're creating a body of trash calling itself science
Is there a single paper that uses a sero prevalence denominator? Is there a single researcher in this field whose brain is working? Just one. That's all I ask for.
Totally wrong. Because Vincent is not thinking about the counterfactual correctly.
*Teachable moment*
1 These drugs were approved by accelerated approval in the LAST line. Some later improved survival & others didn't in an EARLIER line.
2 The counterfactual to AA... 🧵
is demanding RCTs powered for say, OS (survival). If a company couldn't use RR to get AA, they wouldn't run a trial in the 2nd line or 3rd line setting powered for OS, they would run it in the 7th or 8th line
Why? more dire = faster result
3. We have proven that in these v late lines RR and median DOR (the current AA criteria) result just as fast as OS pubmed.ncbi.nlm.nih.gov/30933235/
Many people want covid to be worse than it is. They imagine it has long-term consequences that are worse than other respiratory viruses. Adjusting for severity of illness, it doesn't. Only anosmia is unusual. Why do they want it to be worse than it is? 🧵
For some reason, some people want to live with perennial precautions. They don't want to take off the mask, they want their children to mask, they want to keep getting booster after booster. I don't know why but they want to live in fear.
They have...
A mountain of retrospective observational data that they think supports their claim that covid has long-term disability. That it's a vascular disease. And all sorts of other claims. Nearly all these studies are flawed. They don't have good controls.
Lots of prior studies show many psychological findings don't reproduce. Obviously that's because so much of this science is bullshit. Small sample size, weak methods. Entire fields struggling to justify their existence & people p hacking and exaggerating to be on @HiddenBrain
How do you fix this problem? No one really knows. All of academia is incentivized for hype and discovery, actually being a thoughtful student, criticizing things, pushing for better methods that's unpopular.
Thank you @Erman_Akkus for ur reply it is a good learning opportunity for #ESMO24
In 15 tweets, I will summarize the trial, my criticism, and why this reply contains 3 common errors that oncologists make because our training doesn't teach these ideas.
First, the trial...
The trial is #LEEP-012 and randomizes pts with INCURABLE (see pic) liver cancer to TACE (embolization) plus costly drugs or embolization alone.
These 2 drugs are TOXIC (lenva is horrible) and cost a FORTUNE 200-300k per annum per person
#ESMO24
Every single person has the cancer return. It is non-metastatic, as @Erman_Akkus says, but it is not curable.
Here is the time until measured lesions grow 20% or new lesions present or the patient dies
That's what he and others are excited by #ESMO24
a 4 months PFS
Just out on @medrxivpreprint and @CityJournal
We analyzed all COVID19 corrections in the @nytimes
We show that 2:1 the Times OVERSOLD covid risk or FALSELY HYPED the benefit of restrictions
1 reporter alone was 7% of all corrections
Let's take a look 🧵
Naturally, in times of crisis, newspapers won't get everything right
Corrections are inevitable and forgivable
But corrections should occur at random
Sometimes they are too high and sometimes too low
If they tend to be in one direction...
That's bias
Here is the key figure
When the New York Times made mistake those mistakes were TWICE as likely to OVERSTATE the harm of COVID, particularly to kids, than UNDERSTATE
This is evidence of systematic BIAS in their reporting