By David Tuller, DrPH
I have recovered sufficiently from my post-election coma to send off another of my irritating letters to journals–this one to The BMJ.
As I mentioned in a post earlier this week, The BMJ has corrected a major paper: “Clinical effectiveness of an online supervised group physical and mental health rehabilitation programme for adults with post-covid-19 condition (REGAIN study): multicentre randomised controlled trial.” (Post-Covid-19 condition, or PCC, designates a specific definition for Long Covid.) The correction, published in May, involved adding to key sections of the REGAIN paper the salient detail that the study sample included only patients who had been hospitalized for Covid-19. As a matter of scientific interpretation, the findings cannot automatically be generalized to include all PCC patients–the great majority of whom have not been hospitalized. Yet in the conclusion of the abstract, as well as in a prominent section called “What This Study Adds,” the investigators omitted any mention of this caveat, leaving the impression that the findings could be broadly extrapolated.
It’s all well and good that The BMJ corrected the paper. But as I pointed out in the earlier post, an accompanying editorial, which had been commissioned by the journal and was not externally peer-reviewed, did the exact same thing. And that editorial has not been corrected. So today I sent the following e-mail to Dr Kamran Abbasi, editor in chief of The BMJ.
**********
Subject: Request for correction to editorial accompanying REGAIN study
Dear Dr Abbasi—
In February, The BMJ published “Clinical effectiveness of an online supervised group physical and mental health rehabilitation programme for adults with post-covid-19 condition (REGAIN study): multicentre randomised controlled trial,” from McGregor et al. A dozen colleagues and I sent the journal a letter expressing concerns about several aspects of the REGAIN paper—including the apparent extrapolation of the findings to everyone with post-Covid-19 condition (PCC) rather than just to the small percentage of patients who, like the study participants, had been hospitalized. Other observers raised similar objections.
In April, the investigators posted a rapid response rejecting the various criticisms. In May, however, The BMJ published a correction. The correction addressed the charge that the REGAIN paper misrepresented the findings by omitting from key sections the salient detail that all the participants had been hospitalized.
The correction is an important and welcome update of the public record, since it clearly limits the presumed applicability of the REGAIN findings to a narrow sub-group of PCC patients. Unfortunately, it was issued long after the media had already touted the study results without making any such distinctions.
The correction raises some new questions:
1. Given that the rapid response from the investigators rejected the need for corrections, why did The BMJ decide to publish one anyway, and did the investigators agree with that decision?
2. Has The BMJ issued a press advisory about the REGAIN correction, or made any efforts to reach out to news organizations that covered the initial publication?
3. Alongside the REGAIN paper, The BMJ published a commissioned editorial called “Rehabilitation for post-covid-19 condition.” As with the paper itself, the editorial’s most prominent sections presented the intervention’s reported benefits without mentioning that this cohort of patients had been hospitalized.
Only in the very last paragraph did the editorial cite the caveat about the study population, noting in passing that “inclusion criteria required a history of hospital admission for covid-19, and it is unknown if findings can be generalised to patients…who do not require admission.” And yet, for all intents and purposes, the editorial’s preceding paragraphs did just that—generalize the REGAIN findings.
The REGAIN paper was corrected for having failed to highlight an obviously pertinent point in a sufficiently prominent manner, so it is perplexing that the editorial, which has a similar flaw, has not also been corrected. The critical information about the study sample, which indisputably impacts the interpretation of the findings, should have been provided to readers at the beginning of the editorial–not tucked away at the end like an afterthought. Under the circumstances, does The BMJ plan to correct the editorial as well?
Thank you for your attention to this matter.
Thank you for your attention to this matter.
Best–David
David Tuller, DrPH
Senior Fellow in Public Health and Journalism
Center for Global Public Health
School of Public Health
University of California, Berkeley