CroatMedJ 51 0007
CroatMedJ 51 0007
CroatMedJ 51 0007
doi: 10.3325/cmj.2010.51.7
London, UK
3
Journal of Epidemiology & Community Health, Barcelona, Spain
Discussion Group, April 2009 4
Esteve Foundation, Barcelona, Spain
5
Gaceta Sanitaria, Barcelona, Spain
6
BMJ, London, UK
7
Croatian Medical Journal, Split, Croatia
8
Norwegian Electronic Health Library, Oslo, Norway
*Authors are listed in alphabetical order, except for H. Marcovitch,
chairman of the discussion group and past chairman of COPE.
A systematic review and meta-analysis suggests that Dr Virginia Barbour, editor of PLoS Medicine, raised the fol-
around 2% of scientists admit to have falsified research at lowing issues:
least once (1). Up to 33% admit other questionable prac-
tices such as plagiarism, duplicate publication, undisclosed – Does past or present employment with industry preju-
changes in pre-research protocols or dubious ethical be- dice authors and researchers?
havior (1). There can be no doubt that discovered cases of – Journal policies on competing interests are often un-
research and publication misconduct represent a tip of an clear or ambiguous even though the problem is common
iceberg and many cases go unreported (2). and consumes a disproportionate amount of editors’ time.
– It is difficult to persuade academic institutions and indi-
Experienced biomedical journal editors are aware of a vidual researchers to comply with accepted definitions of
“rogues’ gallery” of major fraudsters, such as Schoen,
Hwang, Sudbo, Poehlman, Singh, and Chandra (3-8). Much
more common are the less dramatic, because more subtle Figure 1.
but probably more dangerous, examples; these are more
dangerous because they remain undiscovered so may
feed into meta-analyses and guidelines.
www.cmj.hr
EDITORIAL Croat Med J. 2010; 51: 7-15
misconduct. tures may represent respect for one’s seniors rather than
– Editors and publishers have not got to grips with non- theft of their work.
financial conflicts.
– There is uncertainty and inconsistency about sanctions In any case, detecting plagiarism is easier for editors of
for misconduct. large and well-resourced journals that can afford to employ
staff to use software for detection. For the overwhelming
Magne Nylenna, editor-in-chief of the Norwegian Electron- majority of journals plagiarism detection is based more on
ic Health Library, raised the following issues: chance than on forensic detection.
– Good practice vs misconduct is not a matter of black In whatever manner COI presents, we need internation-
and white. There are shades of gray. ally accepted definitions: thus, “corruption is the abuse of
– Of greatest significance is the researcher’s/author’s in- entrusted power for private gain” (9) should be axiomatic.
tent – and that may be difficult to determine. There are three aspects to this definition, all culturally de-
– Extreme cases are probably unpreventable so, as in pre- pendent –private gain (eg, journals and reprints, citations
ventive medicine generally, attention to the whole popu- in deciding promotion); power (whom are you serving?);
lation is likely to have a better outcome than concentrat- and abuse, such as cultural behavior deemed improper by
ing on outliers. others. At the symposium, an example was given of a prin-
- Simple guidelines are more useful than detailed, lengthy cipal investigator who, faced with possible misconduct by
protocol a researcher, sent a junior researcher to investigate with
– Everyone concerned merits training. neither the knowledge nor the authority to be sufficiently
critical. The choice of this researcher was, however, in line
Ana Marušić, co-editor in chief of the Croatian Medical Jour- with local norms.
nal, addressed the specificities of small journals:
Science’s attitudes to COI
–There are particular problems for small journals whose
editors are likely to be academics, not full time medical Participants considered that the scientific community is
journalists. largely skeptical of the frequency and adverse impact of COI.
– There is special risk from “scientific inbreeding” (the There is a commonly held myth that science is “self-correct-
tradition of having tenure in the same institution where ing” and that, by default, scientists behave ethically. In addi-
one trained). tion many scientists fail to recognize that they have compet-
– In post-communist countries the concept that it is ing interests so, unsurprisingly, do not declare them.
praiseworthy to cheat the government may be extended
to other perceived authorities. Editors and publishers may be more cynical but are still less
– There needs to be a clear definition of when and than properly alert to COI, especially where it is not finan-
how an article is retracted and what should be the cial. Industry connections raise suspicions but should not
consequences thereof. government appointments do the same? It was pointed
out that in certain disciplines, for example toxicology, there
Specific issues discussed is a regular flow of personnel between government regula-
tors, industry, and academia. Of course, publishers and edi-
Generalizability tors might also stand to gain from having COI, in terms of
enhancing circulation, obtaining finance for reprints or just
We all need to be aware of cultural differences between improving the impact factor
countries or even between academic disciplines in such
matters as ethics and law, as they are applied to science. Motives
While science attempts to be universal and objective, issues
which lead to research contamination, such as through The “Publish or perish” principle dominates researchers’ and
conflicting interests, are essentially local and variable. authors’ behavior (10). Modifying this behavior will require
widely accepted policies and adequate training and men-
Whether or not it results from a conflict of interest, a toring and incentives other than just publication as an aca-
prime example is plagiarism, which in certain cul- demic reward.
www.cmj.hr
Marcovitch et al: Conflict of Interest in Science Communication
The need for standards cine, having discovered that in nearly half the articles on
drug therapy published in the journal between 1997 and
Those journals which screen for image manipulation (eg, 1999, authors had an undisclosed financial link to the prod-
PLoS journals) tend to quantify it as about 1% of accepted uct manufacturer (13). Weatherall pointed out that govern-
papers. What is especially surprising is that when authors ments worldwide were pushing universities to link up with
are asked to provide original data to further investigate po- industry and that “reduced support for clinical research is
tential manipulation, some 25% of authors respond that driving investigators towards industry as a source of fund-
they do not have the original data. Indeed there is no stan- ing.” In 2009, Marcia Angell, ex-editor of the New England
dard approach to data storage in academic institutions Journal, quoted US Senator Charles Grassley, ranking Re-
(compared to industry where there are much more strict publican on the Senate Finance Committee, as revealing
standards). We need to have agreements about such mat- that a researcher at Harvard Medical School had received
ters for example how to maintain and enter record of ex- US $1.6 million (€1.1.million) in consulting and speaking
periments and how to archive them. One difficulty is that fees between 2000 and 2007 from pharmaceutical compa-
the accumulation of data during a project does not follow nies (13). These companies included those which manufac-
the neat process of IMRAD (Introduction, Methods, Results tured drugs which he had advocated in numerous journal
and Discussion) but rather tends to be random in time. publications. Another US academic was stated by Senator
Grassley to control more than US$6 million (€4.2 million)
Authorship of stock in a company he co-founded which was testing
a drug used in psychotic depression; at the same time he
It is well recognized that there is poor agreement between was a principal investigator on a US National Institute of
editors and researchers as to the precise definitions of au- Mental Health grant that included research on the drug,
thorship or contributorship. This leads to disputes about about which he had co-authored three papers. Angell esti-
“guest” and “ghost” authors. mated that, in the US, pharmaceutical companies pay tens
of billions of dollars annually to physicians.
Reviewers
Another area of research where COI arguments have be-
Reviewers are often unaware that they have potential COI come prominent is in toxicology. In an overview, Clax-
outside the obvious (ie, financial) – for example, personal ton, from the US Environmental Protection Agency, has
friendship (or enmity), religious beliefs or nationalism may rehearsed the debate (14). For example, there has been
get in the way. A reviewer may be over-enthusiastic or un- criticism that health assessments of some potential car-
necessarily hostile about a topic, without realizing he or cinogens were carried out partly by scientists employed
she is an outlier in that respect. In small subspecialties or by industrial organizations which produced or distributed
communities, finding a genuinely independent reviewer the materials in question. There is, of course, a counter-ar-
may be difficult. gument: namely that industrial scientists’ work is highly
regulated, meets high standards and unfavorable results
In many journals, reviewers’ COI is not addressed. Esteve are not suppressed (15). The subject is complicated by the
Fernández pointed out that in small scientific communi- presence, both in research teams and in those designing
ties, which may have only 2 local journals, the reviewers regulation or guidelines, of scientists employed by gov-
know the authors personally and sometimes do research ernment – which might itself have a vested interest in the
in the same topic, which is in itself COI, which has to be de- data eventually published. Moreover, in many fields scien-
clared and managed. In his experience, only some journals tists move freely between academia, industry, and govern-
allow exclusion of researchers with competing interests ment, sometimes holding down posts funded by more
from reviewing the submitted manuscript. Finally, editors than one institution.
may also have their own conflicts of interest (11).
Claxton (14), in a non-systematic review of toxicological
Financial conflicts and other links with industry studies where conflict of interest was discussed, acknowl-
edged cases where COI proved to have “negatively im-
In a Lancet commentary in 2000 (12), Professor David pacted the designing, conducting, analysis and use of
Weatherall from the University of Oxford discussed the research efforts.” But he also uncovered evidence that
public apology issued by the New England Journal of Medi- such outside influence and collaboration “can be
www.cmj.hr
10 EDITORIAL Croat Med J. 2010; 51: 7-15
More recently, a review of 1534 original oncology papers In the end, open disclosure provides extra information with
published in eight leading journals showed that 29% had which readers, consumers, and providers of grants can de-
conflicts of interest and 17% declared industrial funding. cide on how much credence to place on the data present-
Where a COI was present, survival outcomes of studies fo- ed. While the large majority of researchers, institutions, and
cusing on treatment were more positive (18). companies are honest, secrecy – whether deliberate or ac-
cidental – can breed only suspicion.
Finally, it has to be kept in mind that research is always per-
formed in the social and economical context, which very Shades of gray
often determines the topic of research. This is the reason,
for example, for more research in clinical medicine then From what has been stated above, it is clear that the simple
in public health or health services (19). presence of COI does not imply fraud or even any miscon-
www.cmj.hr
Marcovitch et al: Conflict of Interest in Science Communication 11
Figure 2.
The continuum from honest error, over misconduct to outright fraud (top horizontal axis). The vertical axis shows the continuum from non-intentional to
intentional nature of the publication behavior, as well as from their easy to difficult detection.
duct at all. COI of some sort, especially non-financial, may ply is that he or she did not realize that they might have
be inevitable, so not necessarily culpable. Nylenna and Si- such a problem; sometimes this may be because they have
monsen postulated that there is no distinct line between not read the journal’s instructions to authors or advice to
ethical and unethical behavior; rather there is a continuum contributors. Alternatively it may be because the journal’s
from honest error to outright fraud and what matters is in- requirements are not stated, ambiguous or incomplete.
tention, a quality which may be impenetrable to the ob- Thus, some journals simply require a general statement
server (25). Nylenna has updated and amplified an illustra- that the author knows of no significant competing interest.
tion from that paper which is shown as the Figure 2. Others ask for specific information on funding. A minority
demand completion of more comprehensive checklists,
COI may be part of the problem at any step in the progres- enquiring into the precise role of any supporting institu-
sion shown in the Figure 2, perhaps becoming more likely tion in design, analysis or preparation for publication. Cur-
as one proceeds steadily toward the more serious transgres- rently, many of the larger journals are conducting a trial of
sions. In deciding how to prevent or deal with corruption a standard unified declaration available from the Interna-
of the scientific record, “shades of grey” need to be borne in tional Committee of Medical Journal Editors (ICMJE) (26)
mind. Marušić pointed out at the symposium that there may This declaration (available from http://www.icmje.org/coi_
be an additional category in post-Communist countries. Pre- disclosure.pdf ) requires all authors to complete a 4 page
viously it was widely regarded that cheating the government questionnaire which seeks information about any resourc-
was praiseworthy; in the post-communist era, this mindset es received by the author or his or her institution to enable
may have been extended to other perceived authorities – in- them to complete the work; it also asks for all sources of
cluding employing institutions, funders, and editors. revenue relevant to the submitted work received over the
previous 3 years – not just from the entity which spon-
Journal policies sored the research. Similar information is required of any
similar financial arrangement with the spouse, partner
When an editor informs a submitting author that he has or children under 18 years of all authors. They are also
concerns about a possible undeclared COI, a frequent re- required to report any personal, professional, politi-
www.cmj.hr
12 EDITORIAL Croat Med J. 2010; 51: 7-15
cal, institutional, religious or other associations “that a rea- sidiary to their research or clinical employment and is of-
sonable reader would want to know about in relation to ten unpaid. It is hardly surprising that they are sometimes
the submitted work.” Tick boxes are included specifically blind to problems that are all too evident to their “profes-
relating to gifts, grants, honoraria, payment for manuscript sional” colleagues from major journals. One small comfort,
preparation, patents, royalties, payments for educational of course, is that many small specialty journals publish few
activities, stock options, travel and accommodation re- studies, such as drug trials, where significant financial in-
imbursement, board membership, consultancy, employ- terests may be at stake, so may be less prone to authorial
ment, and fees for expert testimony. misconduct.
It remains to be seen whether authors (and reviewers) Editors cannot claim to be without resources. Advice on
comply with these stringent requirements and whether recognizing and dealing with COI is available from the
journal editors police them adequately. An important cri- Web sites of ICMJE, the World Association of Medical Edi-
terion is that the corresponding author is expected to seek tors (WAME) (www.wame.org), COPE (www.publication-
and confirm information for co-authors at the time of sub- ethics.org) and the Council of Science Editors (www.coun-
mission. This is important given the results of a prospective cilscienceeditors.org) among others. Many publishers,
study on the journal disclosure form for authorship required including Elsevier and Wiley-Blackwell also provide ethical
by the Croatian Medical Journal. Ilakovac et al interrogated advice for their editors. The latter’s position is described
919 authors of 200 papers (27). The corresponding author in detail in its Best Practice Guidelines (30) which makes
of the papers completed checklists for all 919 authors and it clear that editors, authors and peer reviewers all have a
a blank form was subsequently sent to each individual au- responsibility to disclose interests “that might appear to af-
thor. The main outcome measure was test-retest differenc- fect their ability to present or review data objectively.” The
es between the corresponding authors’ self-declarations. guideline authors state that the existence of a COI does not
Reliability of answers between the two requests was low prevent someone being listed as an author although “edi-
to moderate both for corresponding authors themselves tors may prefer not to commission subjective articles (for
and when acting as proxies for co-authors. example editorials or non-systematic reviews) from author
with a COI” but remind editors that adapting such a policy
Not only authors, but reviewers and editors have potential runs the risk of encouraging authors to conceal relevant
COI. How seriously the latter two groups take their policing interests so may be counter-productive.
role will be crucial. Again, the evidence is a matter for con-
cern. In a survey of 37 journals, 19/30 respondents consid- Institutional responses
ered it important or very important that they declare their
editors’ COI; 13/30 gave the same answer about their editori- Systematic data are not available. However, qualitative
al board and 11/30 those of other editorial advisers. Only half data from the COPE suggests that institutions vary greatly
of those who considered the matter important had a policy in their response to allegations of misconduct by an em-
to deal with the issue, which was “internal” and “often vague.” ployee, including that of undeclared COI. Frequently, edi-
Few had mechanisms for updating declarations (28). tors who have been advised by COPE to request an in-
stitution to investigate a matter of concern find that the
More recently, Wager et al received responses from 231 edi- response is tardy or no action results. Nonetheless, many
tors-in-chief of medical, health care, life sciences, and social institutions have clearly defined and publicized policies for
science journals published by Wiley-Blackwell. Offered a per- handling misconduct.
ceived severity score for their journal of 0 (not a problem) to
3 (a very serious problem), the mean rating for undisclosed In general, research institutions or funders are responsible
authorial COI was 0.73. Asked how frequently they encoun- for dealing with misconduct by their employees or those
tered the problem (on a 0-4 scale for increasing frequency) who have contracted with them. In certain countries, re-
the mean result was 0.90. The authors concluded that at least view and appellate functions reside in governmental bod-
some editors of science journals may be unaware of many ies – particularly in regard to government funded studies.
of the ethical problems that they might be facing (29). The most developed national means of investigating mis-
conduct are in the USA, Canada, Scandinavia, Australia, and
The vast majority of biomedical journal editors are “am- Germany with fledgling operations in the UK, Croatia, Chi-
ateurs” in the sense that their editorial role is sub- na, India, and Japan (31).
www.cmj.hr
Marcovitch et al: Conflict of Interest in Science Communication 13
Regulatory bodies also vary in their responses: in the UK, vasive error or unsubstantiated or irreproducible data. NLM
the General Medical Council, with whom all practicing does not differentiate between retraction due to honest
physicians must register, declares in its guidance to re- error and that due to misconduct. The citation for the ar-
searchers that: “You must ensure that your judgment about ticle is not removed from the database but is updated to
the research is not influenced, or seen by others to be in- indicate retraction and links the original citation to the
fluenced, by financial, personal, political or other external published retraction notice (34). Thus, there is no specific
interest at any stage of the process. You should always de- reference to COI, although COI may be the prime cause of
clare any conflicts that may arise to an appropriate person, misconduct or error.
authority or organization.” The guidance goes on to specify
to whom researchers must declare any financial matters Elsevier advises its authors that they can withdraw an ar-
– including research ethics committees, research partici- ticle in press or retract one already published for various
pants, nurses and non-clinical staff in the research team. reasons, including infringements of professional ethical
The GMC can investigate complaints that a registered doc- codes, such as multiple submission or duplication, bogus
tor has not complied with these requirements and can authorship claims, plagiarism, fraudulent data “or the like”
apply sanctions, after due process of investigation, which but do not refer specifically to undeclared COI.
might include a ban from conducting research, suspen-
sion or even erasure from the register (32). Before due process by a responsible body has made a find-
ing of misconduct but where the case appears strong, it is
Sanctions are also available to (and used by) the US Office open to editors to publish a “notice of concern” or even an
of Research Integrity in regard to misconduct in research editorial explaining the issues involved – but short of any
or in grant applications that have been federally funded. formal retraction.
There have been calls for regulators to require authors who
are required to submit their work for approval to share un- Conclusions of the discussion
derlying data collected in a study so that others can assess
reliability; this could include agencies posting comprehen- The existence of a COI does not necessarily imply miscon-
sive conflict disclosures on their Web sites along with sub- duct or harm but is well recognized to have the potential
mitted studies (33). to lead to feelings of obligation or reciprocity by the in-
dividual concerned (35). Lack of transparency at the time
Sanctions of submission and publication, when followed by subse-
quent discovery of an undeclared interest, is likely to lead
In general, sanctions are available only to regulatory bod- to a perception by readers and editors that the research is
ies, employers, and sponsors in relationship with research- tainted, even possibly fraudulent. Neither concept is help-
ers and authors. It can be very frustrating to editors if none ful to scientific advance or the public interest in probity.
of these are able or willing to take action. Some editors
have proposed their own sanctions – such as banning au- The contrary argument is that editors habitually conflate
thors from submitting papers for a certain length of time or serious misconduct by academics without corporate ties
warning editors of related journals. But there are potential with far less heinous cases where authors have not re-
legal pitfalls, depending on the particular jurisdiction, of sponded appropriately to unjustifiably over-elaborate re-
what might be alleged as restraint of trade or defamation. quirements for disclosure (36). In an invited polemic, Stos-
In consequence, COPE advises its members that, beyond sel claims that “adverse outcomes objectively ascribed to
rejecting a paper or publishing an expression of concern financial COI are almost non-existent” and that “no evi-
or retraction once a serious allegation has been raised or dence supports that corporate detailing and gifting ad-
proved, they should go no further than asking the study versely affects patient care” and that COI strictures actually
sponsor or employing institution to investigate. inhibit medical advances.
www.cmj.hr
14 EDITORIAL Croat Med J. 2010; 51: 7-15
Rose (37) in relation to disease prevention. Rose defined One. 2009;4:e5738. Medline:19478950 doi:10.1371/journal.
the prevention paradox – namely, a measure that brings pone.0005738
large benefit to the community but offers little to every 2 Titus SL, Wells JA, Rhoades LJ. Repairing research integrity. Nature.
participant means that minor changes in the right direc- 2008;453:980-2. Medline:18563131 doi:10.1038/453980a
tion by most of the population is more effective than ma- 3 Retractions’ realities. Nature. 2003;422:1. Medline: 12621394
jor changes by the few. The Gaussian distribution curve is 4 Cheers for disgraced clone scientist Hwang-Woo Suk who misused
shifted to the left. Nylenna and Simonsen conceptualize Ł400,000. Times OnLine Oct 27 2009. Available from: http://www.
scientific misconduct, which would include deliberate lack timesonline.co.uk/tol/news/science/article6890229.ece#cid=OTC-
of transparency by non-disclosure, as an unhealthy condi- RSS&attr=797093. Accessed: February 11, 2010.
tion diffused throughout the science community to differ- 5 Horton R. Retraction – Non-steroidal anti-inflammatory drugs
ent degrees of seriousness. By moving the whole research and the risk of oral cancer: a nested case-control study. Lancet.
community in one direction, the number of serious cases 2006;367:382. Medline:16458751
can be reduced. This approach requires clear definitions 6 Poehlman ET. Letter of apology for falsification of data. Am J
and simple and readily available guidelines at all levels of Physiol Endocrinol Metab. 2005;289:e357. Medline:16014358
research and practice, including entry, with appropriate doi:10.1152/ajpendo.00218.2005
mentoring and audit; national independent bodies should 7 White C. Suspected research fraud: difficulties at getting at the
be permitted to investigate serious cases and whistleblow- truth. BMJ. 2005;331:281-8. Medline:16052022 doi:10.1136/
ers encouraged and safeguarded. As Marušić pointed out bmj.331.7511.281
at the symposium, any such action is even more cogent in 8 Smith R. Investigating the previous studies of a fraudulent
cultures where “scientific inbreeding” (38) has led to aca- author. BMJ. 2005;331:288-91. Medline:16052023 doi:10.1136/
demics routinely enjoying tenure in the institution where bmj.331.7511.288
they were trained. 9 Transparency International. Global Corruption Report 2007:
Corruption in Judicial Systems. Cambridge: University Press, 2007.
Does it matter? Available from: http://www.transparency.org/news_room/faq/
corruption_faq. Accessed: February 11, 2010.
Editors should not underestimate readers. Many may bring 10 Relman AS. Publish or perish-or both. N Engl J Med. 1977;297:724-
the requisite level of skepticism in considering the validity 5. Medline:895795
or impact of published research but those who are naive 11 Marusic A, Katavic V, Marusic M. bRole of editors and journals
would benefit from training in appraising scientific papers. in detecting and preventing scientific misconduct: strengths,
weaknesses, opportunities, and threats. Med Law. 2007;26:545-66.
They should also be aware of the veneration for science Medline:17970252
that inhabits the popular mind. “It’s in a peer-reviewed 12 Weatherall D. Academia and industry: increasingly uneasy
journal, so it must be true” – whereas scientists know that bedfellows. Lancet. 2000;355:1574. Medline:10821359 doi:10.1016/
all published work is merely part of work in progress, so S0140-6736(00)02211-X
may well be contradicted eventually. In this respect, re- 13 Angell M. Drug companies and doctors: a story of corruption. New
searchers, their institutions, and editors should understand York Review of Books 2009; Jan 15. Available from: http://www.
how they might themselves display a COI when they in- nybooks.com/articles/22237. Accessed: February 11, 2010.
volve the media, for example with press releases. They may 14 Claxton LD. A review of conflict of interest, competing interest
be demonstrating an “abuse of entrusted power.” and bias for toxicologists. Toxicol Ind Health. 2007;23:557-71.
Medline:18717514 doi:10.1177/0748233708089046
Acknowledgment 15 Lexchin J, Bero LA, Djulbegovic B, Clark O. Pharmaceutical industry
The meeting of the discussion groups was organized and sponsorship and research outcome and quality: a systematic
financially supported by the Esteve Foundation (http:// review. BMJ. 2003;326:1167-70. Medline:12775614 doi:10.1136/
www.esteve.org). The opinions of the participants do not bmj.326.7400.1167
reflect the views of the sponsor. 16 Tramer MR, Reynolds DJ, Moore RA. L. Impact of covert duplicate
publication on meta-analysis: a case study. BMJ. 1997;315:635-40.
References Medline:9310564
17 Errami M, Hicks JM, Fisher W, Trusty D, Wren J, Long TC, et al. Deja
1 Fanelli D. How many scientists fabricate and falsify research? vu – a study of duplicate citations in Medline. Bioinformatics.
A systematic review and meta-analysis of survey data. PLoS 2008;24:243-9. Medline:18056062 doi:10.1093/bioinformatics/
www.cmj.hr
Marcovitch et al: Conflict of Interest in Science Communication 15
www.cmj.hr