Plagiarism and punishment

Plagiarism is one of the three high crimes of research fraud. The US Office for Research Integrity (ORI) puts it up there with the big boys, fabrication and falsification, in its definition of research misconduct (http://ori.dhhs.gov). Some have argued that the definition should extend to lesser crimes such as undeclared conflict of interest and duplicate publication, but to my knowledge no one has questioned that theft of another person's work is fraud.

How big a problem is plagiarism? The Committee on Publication Ethics (COPE) lists 18 cases of alleged plagiarism reviewed from 1998 to 2005 (www.publicationethics.org.uk), but as with research fraud generally this is likely to be a substantial underestimate of the true extent. Detection has been difficult in the past, but the internet, which has made plagiarism much easier to commit, is also making it easier to detect, as Michael Cross explains (doi: 10.1136/bmj.39388.668773.47). The BMJ is donating content to CrossCheck, an archive of source material that can be checked against submitted or published work. Once fully developed, this will be a key tool for defending the integrity of the biomedical literature.

What if editors suspect plagiarism? COPE has a series of flow charts that outline what journals should do (www.publicationethics.org.uk), but in the end we rely on academic institutions taking action. All too often they fail. Just over a year ago, the BMJ published an account by Iain Chalmers of repeated plagiarism by an eminent Croatian clinician and academic, Asim Kurjak (BMJ 2006;333:594-6; doi: 10.1136/bmj.38968.611296.F7). A detailed investigation by Croatia's highest body for research ethics confirmed Chalmers' allegations as well as finding numerous counts of duplicate publication. After a joint investigation with the BMJ, the Croatian Medical Journal retracted two of Kurjak's papers that were found to be duplicate publications.

In the face of this evidence, we might have expected swift action from Kurjak's employer, the University of Zagreb School of Medicine. The school's research committee concluded at the end of last year that Kurjak had committed repeated plagiarism, and it promptly referred the matter to the school's Court of Honour. The court's decision is still awaited. Where institutions fail to act against perpetrators of misconduct, science itself is the loser. Unless the dean of Zagreb's medical school, Nada Cikes, shows that she takes Kurjak's offences seriously, the scientific integrity of the whole institution is in question and a cloud will remain over Croatia's research community.

At a meeting in Paris last week, Nicholas Steneck, head of the Office of Research Integrity, emphasised that research fraud is not a victimless crime. Science is subverted, funds are wasted, and patients suffer and sometimes die as a direct result. So why is it not a real crime? If we made it a criminal offence, subject to a country's criminal law, it would have to be taken seriously.