Thread by @elisaperego78

This is the @JAMAInternalMed #LongCovid study, which is drawing huge criticism by top experts because of severe flaws in study design, data collection, and interpretation

I provide 🔽 some comments based on key scientific evidence available on Long Covid

🧵

#apresJ20

https://twitter.com/JAMAInternalMed/status/1457754791560548355

First, we must address conflict of interest. The study's main author is a psychiatrist, as many other authors. Notably, the paper doesn't address key biomedical evidence on #LongCovid
It focuses on illness "beliefs" and suggests cognitive and behavioral intervention as therapy!

We perfectly know that #LongCovid has been proven to be a severe, disabling, and sometimes fatal disease by a significant number of top-notch research papers, including in key scientific journals like @Nature 🔽

It is also formally recognized by the WHO

nature.com/articles/s4158…

Another key problem in the @JAMAInternalMed piece is the use, by the authors, of serology tests to identify previous SARS-CoV-2 infection

Sadly, their use of such evidence is so problematic to call into question their understanding of SARS-CoV-2 antibody dynamics!

The authors in @JAMAInternalMed assume, sadly, that (almost) all people infected with SARS-CoV-2 in their sample will
A. be positive to the single, specific serology test they use (bad)
B. develop antibodies to the virus or are still positive at the time of testing (worse)

As many other scientists have already pointed out
A. a significant percentage of covid patients never appear to make antibodies. This is often proven exactly for those with #LongCovid!
B. SARS-CoV-2 antibodies may also wane over time, or drop below the level for detection

Another issue is the kind of serology test the authors use, and how the data obtained are used to frame interpretation! Their test looks at IgG antibodies against the S1 domain of the spike protein
Basically, they look at antibody response against a tiny portion of the virus only

We know, however, that not all serology tests are of the same quality

We also know that some people might have antibodies against other portions of SARS-CoV-2 e.g. the nucleocapsid (N) but not spike (S)

#LongCovid

This is a case in point: a 26 year old #LongCovid patient developed prolonged symptoms from severe vascular inflammation after a mild initial infection

She was only positive for a nucleocapsid test by Roche, while other serology tests were negative!

ncbi.nlm.nih.gov/pmc/articles/P…

A case study of a #LongCovid patients positive only for nucleocapsid antibodies via a Roche serology test is 🔽

https://twitter.com/elisaperego78/status/1458111717184057355?t=eBwCne5ObqugCqtA_N9XlQ&s=19

The authors also collected dried blood from the patients, by asking the patients themselves to perform blood collection via a home kit, and send it back via mail. I'm not an expert on this kind of procedure. But is it possible errors in data collection and contamination happened?

Given the above, it's impossible to make any real, significant connection (in the way the authors do!) between #LongCovid in their sample and a positive serology test. In other words, it is possible people who were indeed infected don't have a + result to the test administered!

It is, therefore, very well possible that those patients who report to have #LongCovid but have a negative serology test, do have Long Covid! They don't have an "illness belief" as claimed by the authors, but symptoms and sequelae following SARS-CoV-2 infection!