This forum post is hidden because you have chosen to ignore John D. Show Details
This forum post is hidden because you have submitted an abuse report against it. Show Details
Yet another research worker has been shown to have falsified the results contained in many of their published papers. http://www.bmj.com/content/344/bmj.e2019 That this was yet another researcher in anaesthesia-related subject should not be a cause for shame to anaesthetists but for pride, as it is not exclusive to anaesthesia, or even medicine. It’s just that anaesthesia, as so often before, is leading the way in developing scientific medicine.
And in science as a whole, it’s an increasing problem; there were 22 papers retracted in 2001, 139 in 2006, 339 in 2010 and over 200 in the first half of last year. In fact, a whole business has sprung up, documenting, publicising and even selling T-shirts about retractions! But anaesthetists need not feel so bad; the Retraction Watch website lists only 12 anaesthesia related retractions so far. http://retractionwatch.wordpress.com/category/by-subject/clinical-study-retractions/anesthesia-retractions/ Thanks to Dr. Fujii we can expect a lot more.
So if fraud is an endemic problem, it must be removed, and there have been several ways in which the fraudsters have been exposed. Some have had their results found impossible to replicate, leading to closer examination of their primary work. Some, like Joachim Boltd forgot to get ethical committee approval for their made-up studies, leading to closer inspection and discovery. Some have been so lazy as to use identical graphs, tables or illustrations in several papers, which some sharp-eyed and well-read academic noticed and questioned.But those can only be used after publication. It is clearly a responsibility of journals and the referees that carry out peer review to weed out and expose fraudulent papers, but they need some tools to use, and they are available. A little known statistical phenomenon is called Benson’s Law which Ben Goldacre has described in his Bad Science articles better than I could do. It is about the frequency of the first digit in a group of large numbers, and that a ‘1’ should be more common than a ‘2’, and so on. http://www.guardian.co.uk/commentisfree/2011/sep/16/bad-science-dodgy-stats
The use of this method has been validated: http://www.ncbi.nlm.nih.gov/pubmed/22695776
Other statistical methods are available, many developed in the banking industry to detect bad accounting as well as financial crime. http://videolectures.net/mmdss07_hand_stf/
But there must be a responsibility on the colleagues of authors, especially if they are invited to be co-authors to prevent such papers being submitted at all. Stories of professors who have their name above a paper who have never set foot in the relevant laboratory are legion. There is a case for them to take some credit for work done in their departments when a professor can be so busy with fund raising, committee and administrative work, but to be a joint author implies that the work was done under their supervision.. There is a need for “research governance” in any department that publishes, in which every member should be signed up to keep to accepted rules of scientific conduct, and to have their own work checked to see that it is so. Neville Goodman, another anaesthetist and researcher, proposed this eight years ago: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1079486/ The member of such a department with that role should change regularly, say every three years. That ‘governance’ role should be to inspect lab notes, equipment and results to detect and stop the struggling researcher before they actually stoop to fraud, if only to get the papers out that they are under so much pressure to produce.