From: 23x11.5c@gmail.com   
      
   Deceit and fraud in medical research   
   Usman Jaffer, , Alan E.P. Cameron   
      
   doi:10.1016/j.ijsu.2006.02.004   
   Get rights and content   
   Under an Elsevier user license   
   Open Archive   
      
      
   Abstract   
   Deceit and fraud in medical research is a serious problem for the credibility   
   of published literature. Although estimating its prevalence is difficult,   
   reported incidences are alarming. The spectrum of the problem ranges from what   
   may seem as rather    
   innocuous gift authorship to wholesale fabrication of data. Potential factors   
   which may have promoted fraud and deceit include financial gain, personal   
   fame, the competitive scientific environment and scientific hubris. Fraud and   
   deceit are difficult to    
   detect and are generally brought to the fore by whistleblowers. Although most   
   cases may be dealt with at an institutional level, regulatory organisations   
   such as the Committee on Publication Ethics (COPE) and Medical Research Ethics   
   Committee (MREC) have    
   been established to monitor and try to remedy the problem.   
      
   Keywords   
   Fraud; Deceit; Medical Research; Medical Ethics   
   Introduction   
   Fraud and deceit in medical research corrupts the scientific record and leads   
   to false conclusions. It leads to a loss in public trust in medical research   
   and doctors. Fraud in medical research affects all grades of investigator.   
      
   Perhaps the most famous case in Britain was that of Malcolm Pearce, a senior   
   lecturer at St George's Hospital Medical School and Geoffrey Chamberlain,   
   Professor and head of the department.1 In 1994 they published a paper in the   
   British Journal of    
   Obstetrics and Gynaecology, of which Pearce was assistant editor and   
   Chamberlain editor. The paper claimed success in re-implanting an ectopic   
   pregnancy. In the same issue, Pearce had published a randomised controlled   
   trial. A few months later, a junior    
   researcher in their department alerted authorities that the re-implantation   
   case was a work of fiction and that the patients randomised in the trial did   
   not exist. The affair led to Chamberlain resigning and Pearce being struck off   
   the medical register.    
   The story warranted front page attention in the Daily Mail newspaper.   
      
   Although this was a rather sensational case, fraud and deceit in medical   
   research has a 'pathogenesis' and manifests itself in a number of ways. In   
   this paper the nature and causes of fraud and deceit in medical research are   
   examined as well as potential    
   solutions to the problem.   
      
   Definitions of fraud and error   
   The committee on Public Ethics (COPE) holds that the principle underlying   
   misconduct is intention to cause others to regard as true that which is not   
   true.2 Misconduct, therefore, not only involves a particular act or omission,   
   but also the intention of    
   the researcher, author, editor, reviewer or publisher involved.   
      
   Although an honest mistake is very different from deliberate fraud, it is not   
   enough to say that all human activity including research is prone to error.   
   The scientific record is equally at risk from honest mistakes as from   
   deliberate fraud. Hence it is    
   the responsibility of the authors to check their work and ensure accuracy.   
      
   Prevalence of fraud and deceit   
   It is difficult to estimate the prevalence of fraud and deceit in the medical   
   literature directly, but an indirect insight can be gained from questionnaire   
   surveys. One North American study reported that 36% of doctoral and   
   post-doctoral students were    
   aware of an instance of scientific misconduct; furthermore, 15% were willing   
   to do whatever was necessary to get a grant or publish a paper.3 Another   
   survey of Biostatistician members of International Society for Clinical   
   Biostatistics, who may be    
   thought of as unbiased and in a prime position to observe fraud and deceit,   
   revealed that 51% of respondents knew of fraudulent projects. Different forms   
   of fraud such as fabrication and falsification of data, deceptive reporting of   
   results, suppression    
   of data, and deceptive design or analysis had been observed in fairly similar   
   numbers.4 This evidence fairly convincingly debunks the idea that fraud or   
   deceit in medical research is a relatively minor and isolated activity. It is   
   indeed widespread.   
      
   Potentially dubious practices in medical research   
   Plagiarism   
   Plagiarism can range from the unreferenced use of others' published or   
   unpublished ideas to submission under 'new' authorship of a complete paper,   
   sometimes in a different language.2   
      
   Forging   
   Forging is the invention of some or all of the research data that are   
   reported, including the description of experiments that were never performed.5   
   Cases of forging have been well publicised in the past.6 and 7 Although   
   forging is the most obvious form    
   of fraud and deceit in medical research, it is by no means the only type of   
   misconduct.   
      
   'Cooking' and 'trimming'   
   Cooking refers to retaining and analysing only those results that support the   
   hypothesis being investigated and ignoring data which may weaken the results.   
   Trimming involves smoothing the irregularities in the data to make the results   
   look more    
   convincing for publication. These offences seem mild in comparison to   
   plagiarism and forging however they provide a first step for the vulnerable   
   researcher down the road of fraud and deceit.   
      
   Misuse of statistical techniques   
   Owing to the complexity of statistical analysis and the difficulty many   
   researchers have with it, fraud and deceit can creep in unwittingly. With the   
   ready availability of menu driven statistical software and lack of   
   understanding of associated    
   hypotheses and assumptions, improper techniques may be employed leading to   
   misleading research.   
      
   Irresponsible authorship   
   Multi-authored papers have become increasingly common over the past 10 years   
   with some having up to 40 authors. Often 'authorship' has been granted for the   
   need to accord recognition rather than due to any direct scientific   
   contribution to the endeavour.   
      
   There is as yet no universally accepted definition of authorship. COPE   
   guidelines state that the award of authorship should balance intellectual   
   contributions to the conception, design, analysis and writing against the   
   collection of data and other    
   routine work. The guidelines go on to state that if there is no task that can   
   be reasonably attributed to a particular individual then the individual should   
   not be credited with authorship.   
      
      
   [continued in next message]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)