Role of systematic reviews in detecting plagiarism: case of Asim Kurjak
BMJ 2006; 333 doi: https://doi.org/10.1136/bmj.38968.611296.F7 (Published 14 September 2006) Cite this as: BMJ 2006;333:594All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
I have read Ian Chalmers’s article (BMJ 2006, 333:594-7) concerning
Professor Asim Kurjak's unethical, professional behaviour in his
representations and publications, which have continued over years.
Professor
Kurjak has been publicly excoriated for his egregious misconduct because
his
behaviour was discovered and could be documented. Unfortunately, he is not
unusual in such behaviour here in Croatia.
I am emeritus professor from a US university who, over the past
decade, has
had personal experiences with members of academic and research
institutions in Zagreb. Without detailing them, I have witnessed numerous
instances of misrepresentation, plagiarism, falsified and fabricated data
in
published papers, and theft of several kinds. Such behaviour is diffusely
and
deeply embedded in the Croatian scientific community and has pressured
many of those who do not subscribe to this behaviour to emigrate, thus
further aggravating the negative selection that has helped to engender the
problem in the first place.
This behaviour will be extremely difficult to reverse. This is all
the more
regrettable because I have many dedicated, correctly-behaving Croatian
colleagues who privately oppose this behaviour but remain silent because
they fear for their positions. Their choice is either to remain silent in
Croatia
or to emigrate.
While I am able to confirm that the problem exists, I cannot write
about all of
its causes, or how to remedy it. First, however, the problem must be
exposed
and admitted, which is my purpose in writing this unhappy comment.
William L. Ragland III
Professor emeritus, University of Georgia
Kaleta b.b.
21400 Supetar
Croatia
Competing interests:
None declared
Competing interests: No competing interests
To the editor:
Chalmers[1] called for “naming and shaming” plagiarists. At least
naming would be beneficial, if only in aid of specificity, which is a lost
virtue in science. The term, “publish,” means to make public. The glory
an author experiences from attribution to him of a published article or
book carries with it a correlative burden of criticism, including any
allegation of plagiarism, if justified. Only the criticized author can
decide whether the criticism evokes shame.
If plagiarism flourished in the absence of its perpetrators’
exposure, researchers’ morale and motivation could suffer, the rate of
scientific progress could diminish and slow development of innovative
treatments could benefit patients less than more vigorous development
might. Anxiety over plagiarism is thus understandable, though plagiarism
is not as immediate a hazard as is poor medical research,[2] which remains
an ongoing scandal but which seems to inspire only complacency.
The first randomized controlled clinical trial (RCT) appeared in a
medical journal,[4] in 1948. By 1951, the number of RCTs had mounted to a
point sufficient to permit calculation of a plausible rate of good vs.
poor research. According to Glantz,[3] among other methodologic critics,
the rate of poor medical research, indicated by the rate of faulty
methods, has been running at a constant rate of about 50% since then.
Ross[5] published the first example of the methodologic critique in
1951. He addressed only the presence or absence of controls, counted 100
articles, cited four and left 96 anonymous. Methodologic critics, e.g.,
Gore et al[6] enumerate errors they find but, with the complicity of
editors, most cite neither the offending authors nor their faulty
articles. Anonymity in methodologic critiques impedes other researchers
from learning to recognize methodological errors in their own work and,
more importantly, prevents medical practitioners from discerning which
treatments are safe and effective for their patients and which may not be.
Twenty-nine years ago, Gore et al[6] claimed that investigators’
statistical errors occurred “at random” and applied statistical maneuvers
to their figures, as if commission of statistical errors were a natural
process, of unspecified mechanism, as if to absolve investigators of any
responsibility and thereby rationalize anonymity:
...We have not identified the papers with errors. We do not want to
pillory the investigators concerned, rather to draw attention to
shortcomings which, plausibly, occurred at random in the 62 reports.[6]
They attempted to bolster their claim of error-randomness in a table
documenting their ingenious gambits:
1. applying a parametric statistical technique, averaging, to numbers of
errors, which are ordinal quantities, to calculate a mean fractional
number of errors or error-categories, 0.76, which cannot exist in nature,
2. applying the Poisson distribution to calculate an expected frequency of
statistical errors,
3. comparing the observed frequency of statistical errors therewith,
4. computing chi square to compare the observed error-frequencies and the
calculated error-frequencies and determine a goodness-of-fit index: 0.84.
Poisson conceived his distribution as an approximation of the
binomial distribution, if the events in question seldom occur:
THEOREM 1: If the probability of success in a single trial p approaches 0
while the number of trials n becomes infinite in such a manner that the
mean mu = np remains fixed, then the binomial distribution will approach
the Poisson distribution with mean mu.[7]
Gore et al determined that authors of 32 of the 62 papers they examined
had committed one or more statistical errors each, 47 errors, in all.
They did not explain why they considered the Poisson distribution
appropriate for such a high event-frequency.
No other author, of whom I am aware, has endorsed concealment of the
identities of criticized authors as explicitly as have Gore et al, yet
such concealment goes on, without explanation.
In aid of consistency, the time may be ripe for a clever statistician
to publish an analysis of plagiarism to show that it, too, occurs merely
as a random event, even less often than does poor medical research, in
response to unspecified Forces of Nature, and to grant statistical
absolution to its perpetrators.
Eric N. Grosch, MD, Largo, FL
eric.grosch@gmail.com
References:
1. Chalmers Iain. Role of systematic reviews in detecting plagiarism: case
of Asim Kurjak. Br Med J. 2006 Sep 16;333(7568):594-6
2. Altman DG. The scandal of poor medical research. Br Med J. 1994 Jan
29;308(6924):283-284
3. Glantz SA. Primer of biostatistics. 4th ed, NY, McGraw-Hill, 1997
4. Marshall G, Blacklock JWS, Cameron C, Capon NB, Cruikshank R, Gaddum
JH, Heaf FRG, Bradford Hill A, Houghton LE, Hoyle J Clifford, Raistruck H,
Scadding JG, Tytler WH, Wilson GS, Hart P D’Arcy et al. Streptomycin
treatment of pulmonary tuberculosis. Br Med J. 1948 Oct 30;ii(4582):769-82
5. Ross OB. Use of Controls in medical research. JAMA 1951 Jan
13;145(2):72-75
6. Gore SM, Jones IG, Rytter EC. Misuse of statistical methods: critical
assessment of articles in BMJ from January to March 1976. Br Med J. 1977
January 8;1(6053):85-87
7. Hoel Paul G. Introduction to mathematical statistics. Fifth Edition,
John Wiley and Sons, NY, 1984
Competing interests:
I am against plagiarism
Competing interests: No competing interests
Editor- We read with interest the article by Chalmers1 on the role of
systematic reviews in detecting plagiarism. Plagiarism should be convicted
according to the motives behind it. A person who is intending to earn
financial benefits through royalty especially by authoring books, as is
the case of Professor Kurjak2, should be more severely penalized than an author writing a review article, and properly citing it as
references. Patent, publish and prosper is applicable for all the
intellectual properties and we should think seriously how to implement
this in the aforementioned context, including the commercial aspects, as
this is a major challenge for all of us involved in the academic
activities. A candid wide scale discussion on the present issue is the
need of the time.
References:
1.Chalmers I. Role of systematic reviews in detecting plagiarism:
case of Asim Kurjak. BMJ 2006; 333:594-596 (16 September).
2.Blaas H-GK. Scientific misconduct. Gynekologen 2002; 15: 37-40.
Competing interests:
None declared
Competing interests: No competing interests
The following is a proposal that may be of some practical benefit. An
organisation could be formed to respond to plagiarism. It would not
require too many resources in the first instance. Journals would have the
option of affiliating themselves to this organisation. Affiliated journals
would then have a duty to report episodes of plagiarism to this body and
the author would be aware of this. The organisation would then simply list
those that had plagiarised, giving them a score corresponding to each
episode of plagiarism. The organisation would have the responsibility of
analysing all subsequent publications of those authors for further
episodes of plagiarism. Authors would have recourse to challenge the
assertions of the organisation. The consequences of the authors actions
could be dealt with by another body and as an example they could be banned
for a set period of time from publishing in affiliated journals.
Competing interests:
None declared
Competing interests: No competing interests
Dear Editor,
Re Chalmers’ article(1) on a case of plagiarism by Professor
Kurjak, this case is obviously plagiarism because Professor Kurjak
reported other’s research results as his own. However, how about a review
article that compiles many results of many researchers? I agree with the
comment of Dr. Alejandro A. Bevaqua(2) that the correct use of ideas
previously developed is not plagiarism, as long as the author of the
previous idea is cited, and the work cited is found in the list of
references. On the other hand, I don’t agree with Andrew D Weeks,(3) that
‘cutting and pasting’ from internet or whatever other source (text books
or journal) is misconduct, as long as the source is cited and quoted
properly, as many journals permit the use of information from the internet,
though for some journals the number of citation is retricted.(4)
Therefore, for journals that do not want to publish plagiarized materials,
a clear definition of plagiarism, and the methods for proper quotation
should be provided.
1.Iain Chalmers. Role of systematic reviews in detecting plagiarism:
case of Asim Kurjak. BMJ 2006;333:594-596.
2.Alejandro A. Bevaqua. Wich is the limit between plagiarism and
right use of literature? Rapid Responses for Chalmers, BMJ 2006;333:594-
596. bmj.com, 20 September 2006.
3.Andrew D Weeks. Using Google to identify plagiarism. Rapid
Responses for Chalmers, BMJ 2006;333:594-596. bmj.com, 20 September 2006.
4.Neville W Goodman. Restricted citations. Rapid Responses for Fiona
Godlee.On the shoulders of giants, BMJ 2006; 333: 0. bmj.com , 15
September 2006
Jeanne Adiwinata Pawitan, M.D., PhD
Department of Histology, Faculty of Medicine, University of Indonesia,
Jakarta 10430, Indonesia.
Editor of Med J Indones
jeanneadiwip@yahoo.com
Competing interests:
None declared
Competing interests: No competing interests
Editor
The British Medical Journal is to be commended for highlighting a specific
instance of academic plagiarism, 1 and in an associated commentary2
countermeasures such as announcements by biomedical journals of
publication of plagiarised work, future raised vigilance of work submitted
by disreputable authors, and disciplinary sanctions by institutions under
whose auspices these authors’ work was conducted.
A substantial harm that may arise from a plagiarised quality
intervention study being included as a well conducted clinical trial in
metaanalyses, thereby exaggerating the magnitude of treatment effect
supported by the plagiarised work. This exaggerated treatment effect may
be so convincing that the studied intervention is either included or
discarded from treatment recommendations. Any ambiguity surrounding the
clinical issue is seen to have been resolved, dissipating the impetus for
further research on an intervention that is, in truth, not convincingly
effective or ineffective.
Plagiarised studies where only minor adjustments are made (such as
that of sample size and results fabricated to maintain internal
consistency and plausibility) would be difficult to detect as
methodologically good studies not only tend to be similar, but are
reported in a standardised manner. Chalmer’s 1 contention that the use of
systematic reviews in peer review to detect plagiarised work may therefore
not be tenable except in instances where a plagiarised work is identically
reproduced from its source.
References:
1.Chalmers I. Role of systematic reviews in detecting plagiarism: case of
Asim Kurjak. BMJ 2006; 333: 594-6.
2.Roig M. Ethical writing should be taught. BMJ 2006; 333: 596-7.
Competing interests:
None declared
Competing interests: No competing interests
Dr Chalmers's article highlighted a very important subject.
I graduated from Zagreb University School of Medicine in 1989. It is very
disappointing to see how inadequately previous deans (dr. Granic 1991,
dr. Labar 2002) handled the case of plagiarism of Professor Kurjak. It
will be very interesting to see whether the current dean (dr. Cikes) will handle
it differently. Will this dean take away the title professor from Asim
Kurjak, or will the medical
school continue on a path of further losing any normal
criteria, except those of political correctness and blind
loyalty. This tendency is exemplified by a recent case.
A young man had a routine laparoscopic appendectomy performed in
Rijeka University Hospital, Croatia. The laparoscopic surgery ended up
with perforated aorta, and the young man lost his leg and was treated in ICU
for several weeks as a result of this "routine" surgery. One significant aspect of this case is that the Internal Committee from Rijeka
University Hospital, and then
immediately after them the Chamber of Physicians of Republic of Croatia
passed the judgment that the outcome of this surgery was a
statistically acceptable complication, adamantly claiming that there
was no error or malpractice in handling this patient's illness.
The Zagreb University School of Medicine did not voice any opinion on this
subject, as if it was somebody else's problem, not theirs. When the acceptable complication of appendectomy becomes leg
amputation in a country, one would think that the leading medical school in that
country would voice some sort of opinion. One would hope that just
one of all these professors of ethics, internal medicine or surgery
would say something.
Competing interests:
None declared
Competing interests: No competing interests
Tom Jefferson pointed that authors of systematic reviews are in the
unique position to detect plagiarism as well as data manipulation and
fabrication. I believe that they - authors - can and must do more. In our
review [1] we found the case of obvious fabrication of data, but opted
just for the description of some details to explain the reason for
exclusion of this study. One searching the Cochrane Library will find
number of similar 'side findings' by reviewers. Unfortunately most of
these findings (or all?) do not progress to actions against plagiarists
and fabricators.
Is it now a time to recognise that public actions after such findings
is a moral obligation of a reviewer? If yes, how can we simplify the
actions, how to lower the threshold for such actions?
1. Vlassov VV, MacLehose HG. Low level laser therapy for treating
tuberculosis. Cochrane Database Syst Rev 2006;(2):CD003490
Competing interests:
None declared
Competing interests: No competing interests
Editor-
Jefferson states that redundant publications represent misconduct. This
may not always be the case as illustrated in the following example.
In 2001 I offered a review article on late angioedema associated with
ACE inhibitors to several primary care journals, but they expressed no
interest. My paper was then rejected from a small readership journal with
explanation that the journal's primary interest is in publishing original
research. I then submitted it in abbreviated form (correspodence section)
to American Family Physician (AFP). After almost three months of waiting
for response, I asked to withdraw my submission, but was immediately
informed that my article is accepted (although I may withdraw it). By
then, I was convinced that my review deserves publication in a journal
with a large international readership and I submitted it to The Lancet who
accepts "any article that advances medical science or practice". I
informed the Lancet editor that this is a redundant publication and
offered a copy of almost identical version already accepted by the AFP.
The Lancet accepted my paper and published it before AFP (1,2).
I was pleased to see that The Lancet has since added section on
adverse drug reactions. The magnitude of late side effects of medications
is impossible to assess in randomized trials and most information comes
from case reports and retrospective reviews. In these cases, redundant
publications may be desirable, providing that authors disclose this
information, as the goal is to inform and educate large readership.
1.Pavletic AJ. Late angio-oedema in patients taking angiotensin-
converting -enzyme inhibitors. Lancet 2002;360:
493-4.
2.Pavletic A. Late angioedema caused by ACE inhibitors underestimated. Am
Fam Physician 2002;66:956.
Competing interests:
None declared
Competing interests: No competing interests
Professor Kurjak
Dear Editor of the British Medical Journal,
The recently published editorial signed by Iain Chalmers on Professor
A. Kurjak has been discussed extensively among many individuals and
societies in Spain. We have decided to write to you because according to
us you have published several unjustified attacks regarding the work of
Professor Kurjak.
Professor Kurjak has developed intensive scientific and educational
activities in cooperation with several universities and has helped to
educate various Spanish societies. Moreover, he has published together
with Spanish authors several original scientific papers, he has written
many chapters in books, as well as five books in the Spanish language
together with Spanish co-authors. In all of these activities he has proved
that he is a highly esteemed colleague, an excellent teacher and an
outstanding director of Donald School of Ultrasound.
We have never been able to find any criticism from our side on his
extensive cooperation with us; therefore we feel that it is our duty to
remind you of the fact that you have published many in justifications
about Professor Kurjak.
We would like to remind you that BMJ has chosen to publish Iain
Chalmers´ editorial which relates to a heavily criticised publication of
Asim Kurjak from 1974. We would like to assert that the appropriate action
has been taken several years ago; therefore we believe it is not fair to
refer back to things which have become bygones because the matter was
solved a long time ago.
We are aware of the fact that more recently a chapter in a book by
Professor Kurjak contained extensive reference to the pioneering work of
Dr Blaas. However, four references were properly acknowledged, but not all
paragraphs related to Dr Blaas´ work were properly accredited.
Nevertheless, the appropriate action has been taken by the relevant
societies, and the author has apologised for his error of judgement;
therefore we believe that it is not fair to consider this error of
judgement as something which was done on purpose.
Moreover, Chalmers article does not provide any allegations
concerning the publications Professor A. Kurjak and his co-workers in
international peer review journals. In none of these publications any
allegations have been made by the reviewers, neither any subsequent
correspondence has been found which questions the appropriateness of the
used methodologies in the presented results.
We are deeply convinced that Professor Kurjak has done so much for so
many of us in Spain and all over the world in his position as one of the
leading persons in the field of ultrasound diagnostics and perinatal
medicine.
Regards,
Prof. JM Bajo-Arenas. President of the Spanish Society of Obstetrics
& Gynaecology
Prof. JM Troyano-Luque. President of the Spanish Society of
Ultrasound in Obstetrics & Gynaecology.
President of MED-UOG. (Mediterranean Association of Ultrasound in
Obstetrics & Gynaecology, 23 Mediterranean countries, including
Jordan and Portugal)
Competing interests:
None declared
Competing interests: No competing interests