Nearly four years after an analysis of more than 160 papers by Yoshitaka Fujii concluded the chances the data were authentic were infinitesimally small, the British Journal of Ophthalmology has decided to formally retract one of the papers included in that review.
The name Yoshitaka Fujii should ring a bell — an alarm bell, in fact — for our readers. He’s firmly listed in the number one spot on our leaderboard, with more than 180 retractions.
The recently retracted paper — “Ramosetron compared with granisetron for the prevention of vomiting following strabismus surgery in children” — has been included in that retraction total for years, because it was part of a seminal 2012 analysis by J.B. Carlisle that put the odds of data occurring naturally in some of Fujii’s papers at:
… a probability of fewer than one in a decillion (or 1 in 1 000 000 000 000 000 000 000 000 000 000 000), the chance of selecting one particular atom from all the human bodies on earth…
Now, nearly four years later, the BJO has decided to pull the paper, listed in Appendix S1 of Carlisle’s analysis, for entirely unsurprising reasons:
Notice of formal retraction of article by Dr Y. Fujii. This article is being retracted as a result of:
(1) Overwhelming evidence of fabrication, related to the fact that the distributions of many variables reported by Dr Fujii in these studies could not have occurred by chance; and
(2) The inability of Dr Fujii’s institutions to attest to the integrity of the study and/or its data conducted under their auspices, as set out in the Joint Editors-in-Chief Request for Determination of April 9, 2012.
The paper has been cited 11 times, according to Thomson Scientific’s Web of Knowledge.
So, our question was — what took so long? We got an answer from Henry Spilberg, associate publisher at BMJ (which publishes BJO):
The editors and publisher took the decision to do this following a query from a reader late last year which alerted us to Carlisle’s 2012 analysis of the Fujii publications and the fact that a BJO paper was among them. Having read this and discussed in depth we agreed there were reasonable grounds to retract the discredited paper in line with COPE guidelines.
Meanwhile, we also stumbled across another retraction from 2015 of a paper that includes Fujii in its co-author list and was flagged in the final report of the investigating committee of the Japanese Society of Anesthesiologists. The paper –“Nicorandil accelerates recovery of neuromuscular block caused by vecuronium“– is listed on page 36 of the report.
Again, unsurprisingly, the retraction note in the Canadian Journal of Anesthesia cites “overwhelming evidence of fabrication” as the reason:
Further to the Expression of Concern posted online, it is with considerable regret that the Canadian Journal of Anesthesia hereby retracts the above-cited article by Dr. Yoshitaka Fujii as a result of: 1) overwhelming evidence of fabrication relating to the fact that the distribution of many variables reported by Dr. Fujii in this study are exceedingly unlikely; and 2) the inability of Dr. Fujii’s institution to attest to the integrity of the study and/or the data conducted under its auspices, as set out in the Joint Editors-in-Chief Request for Determination of April 9, 2012.
Hilary P. Grocott, MD
Canadian Journal of Anesthesia
This journal did issue an Expression of Concern about the paper (and all others co-authored by Fujii in the journal) in March, 2012. So, we asked the journal the same question: Why did it take so long to retract the paper?
Grocott told us the delay was due to an investigation, and the journal would take a second look to ensure there were no other lingering CJA papers by Fujii that needed retracting:
My current understanding is that all of the [Fujii] article retractions from the CJA have now been completed. Furthermore, my understanding of the situation at the time (as I have only been the EIC the past 2 years) was that the delay with the article in question being retracted resulted from additional time needed to fully complete the investigation by all the anesthesia-related journals involved.
I am looking into all of this in order to ensure that the above is as accurate and transparent as is possible and I appreciate your patience as I complete my own review of the numerous documents involved that preceded my tenure.
Carlisle’s paper listed 168 studies; the report from Japan concludes that a total of 171 were fabricated. Our next project: Figuring out how many of those haven’t yet been retracted.
Update 2/4/16 9:46 a.m. eastern: We received an update from Grocott:
I’ve re-looked into the CJA Fuji articles to make sure everything has been accounted for.
The CJA published a total of 39 Fuji articles.
5 were found to be legitimate in Sept 2012
17 were retracted in Dec 2012
17 were found to be indeterminate, of which 16 were eventually retracted in June 2013
The final one (i.e., from our indeterminate list) was eventually retracted in Aug 2015. This is the one that you are writing about and the delay related to it indeterminate status and the lengthy investigation that was required. The bottom line is that with so many article in the mix, it simply got missed after it had been found to be in need of retraction. After I became EIC in January 2014, I was eventually notified by one of the other anesthesia journal editors (who had been looking into the Fuji retraction issue) that a formal retraction had not yet been published to which I immediately published one. We currently have no other papers (Fuji or other authors) that are being investigated, but we remain vigilant.
Hat tip: Rolf Degen
Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, and sign up on our homepage for an email every time there’s a new post. Click here to review our Comments Policy.