Role of systematic reviews in detecting plagiarism: case of Asim KurjakBMJ 2006; 333 doi: https://doi.org/10.1136/bmj.38968.611296.F7 (Published 14 September 2006) Cite this as: BMJ 2006;333:594
- Iain Chalmers (), editor1
- Accepted 26 June 2006
An editorial published in Nature two years ago noted that journals and universities do not always respond appropriately to plagiarism.1 The case study reported here shows how these institutional failings can lead to recurrent plagiarism and how institutions and journals can help to reduce it.
During a search for studies that might be eligible for inclusion in a systematic review of controlled trials of epidural analgesia in labour in the late 1980s,2 I identified a paper by Asim Kurjak and John Beazley published in Acta Medica Iugoslavica.3 Well over half of the text and some of the data in this paper were identical to material in an unacknowledged paper published three years earlier by other authors in the Journal of Obstetrics and Gynaecology of the British Commonwealth.4 In correspondence, I learnt that these authors had not been contacted by Professors Kurjak or Beazley.
Professors Kurjak and Beazley had both worked at Queen Charlotte's Hospital in London, but the paper purported to be a report of a clinical trial done in Croatia. I first wrote to the British author, Professor Beazley. In a letter sent to me on 25 February 1991 Professor Beazley expressed his surprise and dismay because he had never seen the paper bearing his name. He requested an explanation from Professor Kurjak, now professor of obstetrics at the University of Zagreb, who wrote to me on 26 February 1991 confirming that Professor Beazley had not been involved in the paper. Professor Kurjak's letter to me made no comment on his obviously plagiarised text.
In the report of our systematic review of epidural analgesia in labour, my coauthor and I stated that we had excluded the Kurjak and Beazley article “because it contained long passages of text, and some data, which were identical to material published three years previously in an unacknowledged article by different authors.”2
Because Professor Kurjak had failed to explain the plagiarism, I reported our findings to those whom I thought should investigate it. The editor in chief of Acta Medica Iugoslavica, Nikola Perš, writing from the Croatian Academy, concluded his letter to me:
As a colleague and psychiatrist I believe that the stated problem should be solved in the way which would not harm (the) professional and scientific reputation and respect which Prof. Kurjak has earned in 16 years since the problem paper has been published. (10 June 1991).
I wrote to Stojan Knećeviž, Professor Peršić's successor as editor of Acta Medica Iugoslavica (now renamed Acta Medica Croatica), requesting a copy of any notice placed in the journal to draw attention to the plagiarism, but I have never received any response to my request.
Because Professor Kurjak was the director of a World Health Organization collaborating centre in medical ultrasonography, I also reported my findings to WHO. WHO reacted by withdrawing sponsorship for an international conference on ultrasonography that Professor Kurjak was organising; transferring the meeting of the editorial board of the Manual on Ultrasound Diagnostics from Zagreb to Geneva, and excluding Professor Kurjak from it; and suspending him from the directorship of the WHO collaborating centre for a year, while the centre was monitored (Marsden Wagner, European Regional Office of WHO, personal communication, October 1991).
Finally, I also reported my findings to the dean of the medical school of the University of Zagreb, Mate Granić. He appointed a special committee to look into the matter, consisting of the vice-deans for science, graduate courses, and external studies. His letter to me on 14 June 1991 reported:
The members of this committee have immediately realized the impact of Dr Kurjak's early days mistake on the universal scientific principles. However, we are pleased to see that Prof. Kurjak is fully aware of his mistake…Since Prof. Kurjak is [a] very distinguished expert in the field with significant contributions to our ultrasound medicine, we would appreciate your tactful handling of this case.
I now greatly regret having acquiesced in this request for tactful handling of Professor Kurjak's misconduct. Four years ago it was discovered5 that Professor Kurjak had plagiarised material from a Norwegian PhD thesis6 and published it as a book chapter coauthored with a Croatian colleague.7 A copy of the thesis had been given to Professor Kurjak two years previously by its author (H-G Blaas, personal communication). After the publishers of the book had been informed of the plagiarism, they stopped distribution of the book and republished it without Professor Kurjak's chapter. Professor Kurjak and his coauthor did not deny the accusations of plagiarism but tried to play down their “errors of judgment” (H-G Blaas, personal communication).
The plagiarised Norwegian author and his PhD supervisor (Blaas and Eik-Nes) informed the executive committee of the International Society for Ultrasound in Obstetrics and Gynecology, which decided that Professor Kurjak and his coauthor would be “ineligible for membership of the Society and associated benefits for a further three years” (Karel Maršál, personal communication). The Norwegian investigators also reported Kurjak's plagiarism to the then dean of the medical school at the University of Zagreb, Boris Labar, in March 2002 but they have not so far received any response.
The example of Professor Kurjak's plagiarism that I identified occurred more than a quarter of a century ago. Thanks to an astute referee of an earlier draft of this article, the plagiarism turns out to have been even more blatant than I had thought. In his comments, the referee, Jim Neilson observed:
The 1974 Acta Med Iug paper…is clearly an amalgam of two papers—one of which is, as pointed out by Iain Chalmers, the 1971 paper by Noble et al in J Obstet Gynaecol Br Cwlth 78: 559-63. I have done a little detective work and the maternal acid-base work has been lifted from Pearson and Davies, J Obstet Gynaecol Br Cwlth 1973;80: 218-24. As with the Noble paper, large parts of the text have been used verbatim, with little modification, and with no acknowledgement of the Pearson and Davies paper…The figures in the tables have been modified slightly from both original papers, in the Kurjak and Beazley paper—so this is not only plagiarism, it is also scientific fraud.”
More than 15 years have passed since I detected Professor Kurjak's plagiarism and reported it to the relevant authorities. Yet, as the Norwegian case shows, he seems to believe that he can continue to act with impunity in this respect. This continued scientific misconduct raises serious doubt about the extent to which his publications are trustworthy accounts of his own work. Despite this record of scientific misconduct, Professor Kurjak is clearly regarded with great respect, not only within Croatia but also by many obstetricians elsewhere.
The case I have described is an illustration of a more widespread problem. What could or should a whistleblower like me expect from institutions and journals when plagiarism is reported to them. I can illustrate some general principles by itemising my current expectations in this case:
The University of Zagreb should take steps to check or ensure that the 243 publications currently listed in Medline with A Kurjak as an author can be shown to be trustworthy accounts of Professor Kurjak's work
The University of Zagreb should make publicly available, in English, an account of the methods, findings, and conclusions of its investigation
The university should send the results of its investigation to all the journals and editors of books containing publications by Professor Kurjak, to those who have purportedly coauthored articles with him, and to all the officers of the International Society for Ultrasound in Obstetrics and Gynecology
Journals containing articles of dubious authenticity authored by Professor Kurjak should publish notices drawing attention to these concerns and notify the relevant bibliographic databases (the Medline record (PMID: 4820193) of the 1974 Kurjak and Beazley article published in Acta Medica Iugoslavica has no annotation, only a list of 10 related articles).
This case happens to report instances of plagiarism by a Croatian author, currently a senior member of staff in the University of Zagreb, and involves a Croatian journal. But the case should not be taken to reflect badly on Croatian researchers, Croatian institutions, or Croatian journals in general. For example, the Croatian Medical Journal, which has not featured in any part of this case, clearly takes research integrity much more seriously than most journals do.8
Naming and shaming
Although this is but one case study, it prompts me to make some general suggestions for detecting and reducing plagiarism. Firstly, journals and institutions should heed Nature's editorial call1 to take allegations of plagiarism more seriously than they seem to at present. Secondly, as has been suggested by Tom Jefferson and colleagues,910 journals should use systematic reviews for editorial peer review, as well as considering using software designed to detect plagiarism. Lastly, to reduce the numbers of new as well as recurrent plagiarists, journals, institutions, and professional associations need to expose very publicly those found guilty of this form of scientific misconduct.
Journals and institutions should take allegations of plagiarism more seriously
Journals should use systematic reviews for editorial peer review, as well as considering using software designed to detect plagiarism
To reduce the numbers of new as well as recurrent plagiarists, journals, institutions, and professional associations need to expose very publicly those found guilty of plagiarism
I thank Harm-Gerd Blaas for providing documents showing Professor Kurjak's plagiarism of reports of Norwegian research and, along with Sturla Eik-Nes, Karel MarŠál, Jim Neilson, and Vedran Katavić (the Croatian Medical Journal's research integrity editor), for checking the accuracy of earlier drafts of my manuscript.
Contributors and sources This article is based on the journal articles cited and correspondence with all but one (Professor Boris Labar) of the people named in the text.
Competing interests None declared.