By David Tuller, DrPH
This week I’ve been taking some days for family stuff. But I have a minor victory to report. After a series of e-mails with the British Journal of General Practice about a false statement concerning the cost of “medically unexplained symptoms” to the UK National Health Service, the editor has agreed to make a correction.
The false statement was that MUS among working-age people accounts for 11% of the total NHS budget rather than 10% of the amount spent solely on that age group, as reported in a 2010 study. Proponents for the expansion of programs like Improving Access to Psychological Therapies have repeatedly misquoted the 2010 study, Bermingham et al, in advancing their preferred public policy goals. (To be clear, I am not endorsing the 2010 study’s findings–just expressing concern that the reported findings were inaccurately cited.)
Professor Roger Jones initially suggested that it wasn’t worth bothering with a correction because the referenced data were ten years old–even though his journal had quoted the statistic just two years ago. He also asked me to send in one or two “short sentences” for consideration for publication in the journal. I declined this invitation and pointed out that it was not my responsibility to correct mistakes his journal had published. That responsibility falls solely to the authors and the journal.
Today, after a silence of more than a week following my last message, Professor Jones sent me the following e-mail:
“We are making the correction now – the online version of the editorial will be corrected, and a print correction note will be linked online to the editorial.”
Now that’s not much. But getting these people to back down on anything is a start.
I sent the following note back to Professor Jones.
Dear Professor Jones–
Thank you for your response. You had previously suggested that it likely wasn’t worth the bother because the data were ten years old, so I am pleased you have reconsidered your position. I am also pleased you have taken it upon yourself to issue the correction rather than relying on me to send in one or two “short sentences” to be considered for publication, as you had also suggested.
From my perspective, a proper correction would also include some explanation for how this misreading of Bermingham et al occurred and why it was not noticed by anyone involved in the peer-review and publication process. In addition, I would urge you to include an explanation for why Professor Chew-Graham did not respond to this issue when I wrote her in January and informed her that I planned to seek a correction from the journal. Finally, given the repeated misquoting of this data point by Professor Chew-Graham and other MUS experts, it would be appropriate for BJGP to invite a guest editorial discussing the issue and the potentially harmful impacts of promoting public health policies based on inaccurate numbers.
Given the public health policy implications of this matter, I am cc-ing the original group on my response to you.
David Tuller, DrPH
Senior Fellow in Public Health and Journalism
Center for Global Public Health
School of Public Health
University of California, Berkeley
6 responses to “British Journal of General Practice Agrees to Correction about MUS”
I’ve just looked but I can’t see a correction. I wonder what ‘now’ means in their language.
I’ve just checked online again but I still can’t see a correction.
There’s an oxymoronic phrase that people from Wales sometimes use – ‘now in a minute’ – but we seem to have moved into ‘now in a day’ territory here. Perhaps the correction has been delayed because some thought is being given to what David wrote in his email. I do hope that’s the reason.
‘Now in a week’ ?
Now by the end of the month (June). Yay, success! Thanks for persevering David.
For a couple of days there I thought that the BJGP might do the right thing, ( a bad case of turnip brain, I suspect ). The correction to the online article seems to have been made today, or perhaps yesterday. I don’t know which because I can’t see any note that indicates that a correction has been made to the paper and when the correction occurred. I find it deeply disappointing that the BJGP would not clearly inform online readers at the very least that the article has been corrected and when that correction was made. Why wouldn’t they want readers to know that an error was made in the original article? Who or what are they trying to protect?
David, quite rightly, asked Roger Jones to do a lot more than just the in-text correction to help mitigate the damage from this error, but there’s no sign yet of any of those additional measures being implemented. I’m left wondering whether UK medical journals can be trusted to do anything properly.
The July issue of BJGP is now out and the correction has been listed at the bottom of the online ‘Table of Contents’ page -https://bjgp.org/content/69/684 with a link to this notice -https://bjgp.org/content/69/684/333.4 . There is also a notice flagging up the correction on the online article -https://bjgp.org/content/67/656/106. It seems this has been done in the same way as for corrections to other BJGP articles, so nothing seems to be amiss now and my previous comment of 13th June where I expressed concern that the online article appeared to have not been corrected in the standard way no longer applies.
As yet there is no change to Pubmed versions of the article -https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5325626/, but Roger Jones told David that “Pubmed etc will only make the correction once the print notice has been published” (https://www.virology.ws/2019/06/11/trial-by-error-bjgp-correction-coming-bmj-still-stonewalls/) so we can presumably expect to see the Pubmed versions corrected in the near future.
In my opinion it’s a shame that the BJGP hasn’t done more to help mitigate the potential damage caused by this error, but at least the mistake has finally been corrected by BJGP for all to see.