My short list of infamous anaesthetists [read part one and part two of this blog series] has developed into a musing about research misconduct—and particularly fraud—in general, prompted by the Fujii case, with up to perhaps 200 retractions on their way.
Here, I’m pondering the cost of research fraud: who loses, and who loses more (for nobody really wins, in the end). Some of these themes have been raised by Peter Kranke in a recent editorial in Anaesthesia, in which he reflects on the ten years since he and his colleagues first raised suspicions about Fujii’s work in a beautifully understated letter entitled: “Reported data on granisetron and postoperative nausea and vomiting by Fujii et al. are incredibly nice!” (the italics unfortunately lost in modern e-versions of the reference. It was nice (or even, nice) to be able to offer Peter the opportunity to achieve “closure” in an editorial, a decade later).
First, there’s the cost to the perpetrator, and those close to him. Whilst not suggesting anyone should be too sympathetic for fraudsters themselves, it’s sad nonetheless to see an eminent figure struck down and humiliated, even one I haven’t met personally. I’ve no idea whether Fujii has a family, but presumably he has (or had) friends and colleagues, at least some of whom were unaware of what was going on. Fujii’s predecessors, Boldt and Reuben, travelled and lectured widely before their fall from grace, and were presumably wined and dined accordingly by the great and the good.
Then there’s the shame that might be felt by the institutions, professional bodies, and individuals concerned with regulating the system, and the time, effort, and distress of dealing with the fallout. This can’t be helped by the fact that Fujii’s misdemeanours were apparently taking place for the past 15 years or more, and across several institutions, so the number of people potentially tainted (or who might feel—or be accused of being—thus tainted) must be large.
There’s the cost of reviewing, processing, sub-editing, typesetting, and printing all those thousands of words. With the sad, needs-to-get-out-more curiosity that goes with this job, I’ve counted the number of words in the first five of Fujii’s papers, identified via PubMed, whose full-text versions I’ve been able to access. It comes out as about 1600 words per article—not a lot, though remember we’re talking about approximately 200 retractions here, some of them reviews so quite a bit longer: perhaps a total of a third to half a million words. Kranke estimated at least 3000 hours of processing time. Not to mention the hours and hours (I shall bear the scars for ever) spent investigating, discussing, and eventually retracting the articles.
The one thing in Fujii’s favour is that the number of patients suffering bodily harm as a result of his retracted work is small, despite its volume. Few—if any—will have died, since most of his work was on peri-operative antiemetics, particularly granisetron. (The canine population will have suffered more, since a sizeable minority of his work was in dogs, selflessly giving up their diaphragms and occasionally other organs in the interests of science). Antiemetics are, on the whole, fairly inoccuous drugs, and granisetron is not very widely used—in the UK at least—though there must have been quite a few patients who would have received it but with less benefit than expected, or been deprived of other, perhaps more effective drugs. Perhaps a few might have suffered serious complications as a result, but it’s unlikely. Although far fewer of Boldt’s and Reuben’s papers were retracted, in the areas of intravenous fluids and analgesics, respectively, patients are more likely to have suffered. And of course many thousands of children will have been harmed through not receiving the MMR as a result of just two retracted articles by Wakefield, so the harm isn’t necessarily in direct proportion to the number of retractions.
Sales of antiemetics might have been falsely influenced, which could have resulted in additional, unnecessary costs to hospitals (typically, the new “-etrons” aren’t cheap), and some drug companies’ suffering reduced sales in favour of granisetron. Conversely, the manufacturers of granisetron could have unwittingly benefited, through no fault of their own. Moneys gained and lost may well have been greater for other areas of practice, such as those of Boldt and Reuben.
But the big loser, of course, is trust. Doctors still remain top in the list of trusted professions, whilst “professors” and “scientists” are third and fifth, respectively. Will scandals like the Boldt, Reuben, and Fujii cases dent this record? Or is it mainly damage to trust within the “scientific community” (whatever that is) that’s the main victim? There’s lots of talk currently about increasing regulation in order to prevent and/or detect such scandals, which can only increase the barriers for other researchers, already facing a mountain of regulation. Perhaps, in the end, this will be the main cost we’ll all have to bear, along with the loss of research never done as a result.
Competing interests and acknowledgements: I am editor in chief of Anaesthesia, and a co-signatory of the joint editor-in-chief letters to Boldt’s and Fujii’s institutions. I was a member of the Committee on Publication Ethics Council until earlier this year.
Steve Yentis is a consultant anaesthetist in London and Editor-in-Chief of Anaesthesia.