RETRACTED: Post–COVID-19 Condition in Children, 2023, Hahn et al

[SNT Gatchaman]

Original article: https://jamanetwork.com/journals/jamapediatrics/fullarticle/2809315

Retraction notice: https://jamanetwork.com/journals/jamapediatrics/fullarticle/2822489

Article's conclusion —

The incidence of PCC in this study was strikingly low (0.4%). Most children experienced a resolution of symptoms within 2 weeks of infection. Pre–COVID-19 symptoms were factors in post–COVID-19 symptoms.

Retraction following concerns raised —

We identified a coding error whereby children with missing symptoms data were coded as having no symptoms. This error resulted in 2 participants being misclassified as having symptom resolution when they should have been classified as having PCC. We identified another child who should have been classified as having PCC.

We also identified an error in which we included only participants with at least 2 months of symptom data (coded as 61 days) rather than those with at least 8 weeks (56 days) of symptom data. This difference resulted in the exclusion of 15 participants who should have been included in the analysis. In addition, we originally reported a study sample of children between 8 and 13 years of age. We subsequently identified participants with COVID-19 (cases) who were recruited between 1 and 7.49 years and 14.5 and 19 years of age. These participants were included in the original PCC analysis.

We had concluded that the incidence of PCC was 0.4% (1/271). In correcting these errors, we found that the incidence of PCC was 1.4% (4/286).

 

[bobbler]

interesting. These things seem pretty significant. I wonder how it came about? - are we assuming that someone else spotted it and wrote to them, but then such a reviewer would need to have been able to access the raw data?

I had a quick look at the conclusion/discussion of the original to see what recommendations were made on the basis of the finding.

The incidence of PCC in this study was strikingly low (0.4%). Most children experienced a resolution of symptoms within 2 weeks of infection. Pre–COVID-19 symptoms were factors in post–COVID-19 symptoms.

Strengths of this study include use of pre–COVID-19 symptom data and longitudinal prospective collection of post–COVID-19 symptoms. Limitations include depending on parent-proxy symptom reporting and the narrow age range of participants. Additional research is required into the neurobehavioral sequelae of SARS-CoV-2 infection in school-aged children.

Does the mention of additional research is required into the neurobehavioural sequelae of covid mean they are suggesting that 'as we found none actually got PCC it's a behavioural thing' or am I misinterpreting this?

 

[Amw66]

https://twitter.com/user/status/1825976910859678064


RETRACTION: The @JAMAPediatrics study claiming “strikingly low” incidence of Long Covid (PCC) in kids, retracted due to multiple errors whereby incidence was underestimated by a factor over 2.5. But also it’s methodologically completely invalid anyway.

 

[Amw66]

The retraction letter does not address any of the substantive and invalidating concerns we raised (all the way to the VP of JAMA Network), but instead points to NEW ADDITIONAL errors in coding that invalidate the work. Are the authors trying to save face?

/4

 

[Amw66]

The original thread from Blake Murdoch dissecting the study
https://twitter.com/user/status/1752105777136066715

 

[Amw66]

Another in same vein
https://twitter.com/user/status/1752582003818725649


That's how bad some research on paediatric #LongCOVID is.

In our letter to @JAMAPediatrics we point to obvious & fundamental errors in paper which claims that Post Covid Condition in kids is "strikingly low"

Those errors are so egregious, they should warrant a retraction


1/

 

[bobbler]

The retraction letter does not address any of the substantive and invalidating concerns we raised (all the way to the VP of JAMA Network), but instead points to NEW ADDITIONAL errors in coding that invalidate the work. Are the authors trying to save face?

/4

thanks for the additional info. so there were letters that prompted a closer look to take place.

that seems feasible, I'm also slowly learning that the list of what actually prompts 'retraction' vs doesn't can seem to a layperson somewhat eccentric to 'whether it matters' etc

 

[Amw66]

thanks for the additional info. so there were letters that prompted a closer look to take place.

that seems feasible, I'm also slowly learning that the list of what actually prompts 'retraction' vs doesn't can seem to a layperson somewhat eccentric to 'whether it matters' etc

Sadly the critique is behind a pay wall when the published article wasn't .
Seems like the publishers kind of knew it was poor .

 

[rvallee]

The retraction letter does not address any of the substantive and invalidating concerns we raised (all the way to the VP of JAMA Network), but instead points to NEW ADDITIONAL errors in coding that invalidate the work. Are the authors trying to save face?

/4

They're clearly amateurs and not well-connected, though. Everyone knows that when you get caught like this you treat your critics as hostile, accuse them of all sorts of dirty things, work reporters and editors behind the scenes and abuse positions of authority to get your way.

They could have learned from the pros. Tsk tsk.

 

[SNT Gatchaman]

Just because it was referenced on Bsky today here also is the thread from 10 months ago from Mark Ungrin PhD. This all sounds depressingly familiar @dave30th.

An example of the shenanigans and junk science accepted in medical journals.In September 2023, JAMA Pediatrics published a ludicrously bad paper (quasi-retracted after much effort) claiming long COVID is "strikingly rare" in kids, supposedly vs the WHO definition.

They cited the 2022 WHO adult PCC definition, and claimed the only child in the study who met that definition recovered by 14 weeks after the onset of acute COVID......which is logically impossible: The definition specifies it only "usually" starts within 3 months - and fluctuates and relapses.

That was enough to draw our attention. We realized they appeared to have made it to 2023 and published the study without anyone involved (including reviewers) realizing that "long COVID" does not mean "acute COVID that goes on for a long time".We wrote a letter: jamanetwork.com/journals/jam...

JAMA Pediatrics made us remove any explicit call for a retraction. We assumed that was to give the authors a fig leaf - inappropriate, but consistent with the way senior clinicians are often shielded from criticism. Basic scientific rigour made retraction unavoidable, so we let it go.

Turns out basic scientific rigour wasn't enough at JAMA Pediatrics. They buried our letter behind the paywall (vs original open-access paper) and altered our title after we approved final proofs, without telling us, obscuring that it was critical at all. Altered (L) vs corrected (R).

When they finally admitted it was their fault (we have the emails) they eventually corrected it - and used that as an excuse to add a notice of correction to our letter (but not the response, whose title also changed), worded as if it might have been our fault.What a convenient...accident...

JAMA published a response from the authors admitting they had not even collected the data necessary to assess against the WHO PCC definition (the central claim of the paper) and also misleadingly inserting the 2023 pediatric version (their ref #3 here) - never mentioned in the original paper.

JAMA Peds also allowed the authors to retroactively invent a more convenient definition of PCC, and completely ignore the fact that the original claim was unambiguously "met the WHO PCC definition".NB while JAMA's screwup changed both titles, only ours got a "Correction".

And then JAMA Pediatrics declared the matter closed - with the original, demonstrably, *admittedly* false claims still open access, without even a notice of correction.With gaslighting around PCC from peds docs already a major problem, this was hugely unethical.

JAMA Pediatrics is the world's largest-circulation specialist peds journal, reaching a huge international audience. They published screamingly obvious, harmful, junk. Where was the peds community?Why did it take outsiders to step up?Patients, not colleagues, have to come first.

After being denied access to the data, we'd finally had enough, and working with some great parent / patient representatives we submitted a research misconduct complaint in March 2024.I'm officially forbidden to discuss the findings, but *finally* in August 2024: doi.org/10.1001/jama...

Unfortunately, even the retraction is bad. It fails to cite the fatal flaws we identified, instead claiming to have discovered all-new flaws in the calculations. Written more as a minor math correction, it doesn't even address the original false claim that the study used the WHO PCC definition.

The icing on the cake?By spinning it as just an error of analysis (commonly the work of the first author) the (single-author) retraction throws the (junior, female, trainee) original first author under the bus, and omits the bigger conceptual failure owned by the (senior, male, MD) last author.