Honorary and ghost authorship in high impact biomedical journals: a cross sectional surveyBMJ 2011; 343 doi: https://doi.org/10.1136/bmj.d6128 (Published 25 October 2011) Cite this as: BMJ 2011;343:d6128
All rapid responses
I read with interest the article 'Honorary and ghost authorship in high impact biomedical journals: a cross sectional survey' by Wislar and colleagues, which has already been cited 137 times according to Google Scholar. This is clearly influential work and is a topic that evidently is in the conscience of many of those who currently work in medical research. ICMJE criteria have been criticised by several commentators as being illogical and unethical, for example (http://www.bmj.com/content/343/bmj.d7192). Indeed, these guidelines have been criticised for being too inflexible, since authorship is a complex topic in which contributions can take many forms and are very difficult to quantify. My own view is that it should be left up to the corresponding and senior authors to make their own judgement on who should be an author, but that the scientific community should discourage the addition of blatant guest authors, that is to say those who have contributed nothing or more concerningly those whose presence is designed to curry favour with the journal or end users - that is to say a substantive conflict of interest is introduced.
The line between those who should be authors and those who should be acknowledged is a grey area in reality and I think the ICMJE guidelines are deficient in this regard. Take for example someone without whose contribution the work could not have been completed, or completed to the same standard, but who fails to fulfil all aspects of the ICMJE criteria.
Examples could be:
a) a technician who conducted specialist experiments
b) a collaborator who provided cell lines, reagents and/or data sets
c) a data entry clerk without whose work there would be no data to analyse
d) a PI of a key recruitment site
e) someone who offered generally supportive comments and advice during the project to inspire and/or guide the direction
f) someone who helped with the writing and/or editing of the manuscript, but had no other involvement
g) a head of department or other senior figure without whose co-operation the required facilities and funding would not have been available
These are but a few examples and each case brings its different complexities.
In that case, what should a diligent and responsible senior author or corresponding author do? The first option is to play by the rules and leave that person off the author list. The second option is to include that person on the author list in light of their genuine contribution to the work. According to ICMJE, that person would be a 'guest author'. In such cases, I do consider this rather judgemental and harsh. No one making guidelines or rules can fully appreciate the intricacies of each research study to which they are to be applied. Therefore, I consider it would be far better to leave such matters of judgement to those directly involved in the project.
Potentially, it is the perceived unfairness of ICMJE authorship criteria that leads to attempts to circumvent them, as described in this article by Wislar et al. In situations where people are seeking to circumvent rules that are perceived to be unfair, using self-report measures faces substantial limitations. Even if responses are anonymous, it is questionable the extent to which people would be prepared to admit activities that might be construed by the powers that be as inappropriate authorship. This is a similar problem to that faced in trying to quantify research malpractice (http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0005738). My anecdotal feeling is that the amount of guest authorship reported in this article is substantially underestimating the real situation if guest authorship is defined as someone not strictly meeting all the ICMJE criteria. My thought is that we will never know and can never know the real prevalence of something such as guest authorship. Since there is the fear of judgement and criticism for perceived inappropriate authorship, many people will not be prepared to be honest and open about this matter. The generation of a culture of fear about such accusations is another downside of the strictness of ICMJE criteria - and it turn it may lead to an increase in ghost authorship as some senior authors will be afraid of someone being interpreted as a guest, so leave off a genuine contributor. Most authors are not setting out to do anything suspect - they are just making their best judgement on who should be an author. It is difficult to see in this current climate how the prevalence of guest authorship could be accurately determined. Asking authors is likely to lead to severe under-reporting, while the same could happen with using contributorship statements, which might be manipulated to ensure all authors appear to meet the criteria when the manuscript is submitted.
Competing interests: No competing interests
The study by Wislar et al.  shows that efforts by journals to promote integrity in authorship practices are helping to reduce the incidence of honorary and ghost authorship, which is good news. While journals are taking important steps to resolves issues concerning authorship attribution, authorship order remains an important concern. Most scholarly studies of authorship and professional guidelines focus on questions concerning appropriate versus inappropriate authorship attribution [2-7]. Authorship guidelines developed by the International Committee of Medical Journal Editors  and the Committee on Publication Ethics  address authorship attribution but not authorship order. However, controversies concerning authorship order remain important.
While there is considerable interdisciplinary variation in the meaning of different positions in the authorship order, in most of the biomedical sciences the most prestigious positions are first and last author. In biomedicine, the first author is usually the person who has done most of the work on the project, while the last author is a senior scientist who supervises the research.[11, 12] Corresponding author designation also has special importance, since the corresponding author usually interacts with editors and deals with request for data and materials.
To handle disputes concerning authorship order, some scientific publications now include a note stating that two or more individuals have contributed equally to the research. The most conform form of this practice is to state that two or more authors at the beginning of the list (i.e. first authors) have contributed equally. In most cases, only two first authors are designated, but sometimes more three or more individuals are named. A paper published in Science in 2007 listed four first authors, for example. In some instances authors in other positions, such as last, are designated as having made equal contributions. The trend toward designating equal contributions by authors has been increasing since 2000.
Since the designation of equal contributions does not change the authorship order appearing on the publication or in an electronic database, co-first authors who are not listed first but sometimes call attention to the co-first designation on their curriculum vitae (CV) to receive credit for their accomplishments. They may even switch the authorship order on their CVs, so that people who are evaluating their work will see that they are a first author on particular publication. This may create some confusion if the reader of the CV checks the reference and finds that it is different from the CV listing.
Is stating that authors have made equal contributions to a paper a useful notation or is it a deceptive practice? According to some, stating that two authors have contributed equally is a sham because individuals rarely make equal contributions to a paper. A careful consideration of the contributions of authors will usually reveal that one individual has done more work than another. The probability that three or four individuals have contributed equally to a paper would seem to be very low. Scientists usually state that two authors have contributed equally, according to this criticism, to appease researchers who both want to be listed in the same authorship position, not to clearly indicate that two people have contributed equally. If the main ethical purpose of authorship designation is to allocate credit fairly and honestly, then stating that two authors contributed equally rarely achieves this goal. It is simply a ruse. Individuals who switch the authorship order on their CVs add to the deception, because the CV should reflect the actual listing on the paper.
Because the designation of equal contributions by authors is a relatively recent phenomenon in scientific publishing, it is difficult to determine whether it is a useful notation or ethically problematic. In theory, it could be a useful designation that allocates credit fairly and honestly. In practice, however, it may be abused by some in order to deal with authorship disputes. Since this designation has merit in some cases, it would be wise to provide researchers with additional guidance on using this designation properly, rather than prohibiting it. Journals should consider developing policies that address the assignment of equal contributions by authors. These policies should indicate the equal contribution designation should be used only when individuals have contributed equally to the research. Individuals who are named as having made equal contributions should affirm in writing that they contributed equally to the research. Nature’s authorship guidelines, for example, state that co-first authors can be designated if they have contributed equally to the work. Guidelines may not solve all of the potential problems related to noting that authors have contributed equally to the research, but they can help prevent scientists from abusing this designation.
Acknowledgments: This paper is the work product of an employee or group of employees of the National Institute of Environmental Health Sciences (NIEHS), National Institutes of Health (NIH). However, the statements, opinions or conclusions contained therein do not necessarily represent the statements, opinions or conclusions of NIEHS, NIH or the United States government.
1. Wislar JS, Flanagin A, Fontanarosa PB, Deangelis CD. Honorary and ghost authorship in high impact biomedical journals: a cross sectional survey. BMJ. 2011; 343: d6128.
2. Claxton LD. Scientific authorship. Part 2. History, recurring issues, practices, and guidelines. Mutat Res 2005; 589(1):31-45.
3. Gaeta TJ. Authorship: "Law" and order. Acad Emerg Med 1999; 6(4):297-301.
4. Street JM, Rogers WA, Israel M, Braunack-Mayer AJ. Credit where credit is due? Regulation, research integrity and the attribution of authorship in the health sciences. Soc Sci Med 2010; 70(9):1458-65.
5. Resnik DB. A proposal for a new system of credit allocation in science. Sci Engin Ethics 1997; 3:237-243.
6. When authorship fails. A proposal to make contributors accountable. Rennie D, Yank V, Emanuel L. JAMA 1997; 278(7):579-85.
7. Flanagin A, Carey LA, Fontanarosa PB, Phillips SG, Pace BP, Lundberg GD, Rennie D. Prevalence of articles with honorary authors and ghost authors in peer-reviewed medical journals.
JAMA 1998; 280(3):222-24.
8. International Committee of Medical Journal Editors. Uniform requirements. Available at: http://www.icmje.org/. Accessed: March 6, 2012.
9. Committee on Publication Ethics. Guidelines on Good Publication Practice. Available at: http://publicationethics.org/static/1999/1999pdf13.pdf . Accessed: March 6, 2012.
10. Riesenberg D, Lundberg GD. The order of authorship: who's on first? JAMA 1990; 264(14):1857.
11. Wren JD, Kozak KZ, Johnson KR, Deakyne SJ, Schilling LM, Dellavalle RP. The write position. A survey of perceived contributions to papers based on byline position and number of authors. EMBO Rep 2007; 8(11):988-91.
12. Tscharntke T, Hochberg ME, Rand TA, Resh VH, Krauss J. Author sequence and credit for contributions in multiauthored publications. PLoS Biol 2007; 5:e18.
13. Akhabue E, Lautenbach E. "Equal" contributions and credit: an emerging trend in the characterization of authorship. Ann Epidemiol 2010; 20(11):868-71.
14. Zeggini E, Weedon MN, Lindgren CM, Frayling TM, Elliott KS, Lango H, Timpson NJ, Perry JR, Rayner NW, Freathy RM, Barrett JC, Shields B, Morris AP, Ellard S, Groves CJ, Harries LW, Marchini JL, Owen KR, Knight B, Cardon LR, Walker M, Hitman GA, Morris AD, Doney AS; Wellcome Trust Case Control Consortium (WTCCC), McCarthy MI, Hattersley AT. Replication of genome-wide association signals in UK samples reveals risk loci for type 2 diabetes. Science 2007; 316(5829):1336-41.
15. Drug Monkey. Co first-authorship is a lie and a sham and an embarrassment to our profession. Available at: http://scientopia.org/blogs/drugmonkey/2012/02/28/co-first-authorship-is.... Accessed: March 6, 2012.
16. Nature. Authorship. Available at: http://www.nature.com/authors/policies/authorship.html. Accessed: March 6, 2012.
Competing interests: No competing interests
I think Lacasse & Leo may still have misunderstood some of the points I was making. Apologies if my last post wasn't sufficiently clear.
"We completely understand why Dr. Jacobs would use the ICMJE guidelines as justification for his definition of authorship."
I'm not attempting to justify my definition of authorship. I don't have a definition of authorship. It is the ICMJE who have a definition of authorship, and I simply note that it seems appropriate to follow it when publishing in a journal that has adopted the ICMJE criteria.
"However, in the medical literature, in many cases the roles have been reversed, and those listed in the acknowledgement section have contributed more to the published article than the authors listed on the byline."
Well, maybe. But that really depends on what you mean by contributing to the published article. If you just mean the writing process itself, then you're probably right. But remember that most papers in clinical research are the final stage in a huge piece of work: planning the study, recruiting the patients, treating the patients, following up the patients, analysing the data, and only then writing the paper. The overall contribution of a medical writer who only picks up the project at the manuscript writing stage is probably tiny compared with that of a clinical investigator who has been involved from the start.
"In the paper by Wislar et al. , they state that an article is ghostwritten if: “An individual who was not listed as an author made contributions that merited authorship.”"
No, they don't. They state that a paper has a ghost author if that condition is met. Ghost writing and ghost authorship are not quite the same thing. I think some of the confusion here comes from a failure to appreciate the difference between the two. A ghost writer is someone who has contributed to a paper, but falls short of fulfilling authorship criteria, and is not mentioned on the paper. A ghost author is someone who does meet authorship criteria and is not mentioned on the paper.
"In contrast, if we understand Dr. Jacobs correctly, he believes that a paper with a medical writer mentioned in the acknowledgement section does not constitute a ghostwritten paper."
Yes, that's correct. I do indeed believe that a paper with a medical writer mentioned in acknowledgements does not constitute ghostwriting.
"However, this is not universally accepted"
As Lacasse & Leo don't seem to accept this definition, then clearly that statement is true. However, I suspect Lacasse & Leo are in a rather small minority. Most commentators whose contributions I have read, such as ICMJE, WAME, AMWA, EMWA, and many high profile journals, accept that a paper written by a medical writer who does not fulfil conventional criteria for authorship is not ghostwritten as long as their role is described somewhere in the paper.
Competing interests: As stated previously
We greatly appreciate Adam Jacobs' recent rapid response  and his clarification about how medical writers view the ICMJE guidelines and their relationship to ghostwriting.
We completely understand why Dr. Jacobs would use the ICMJE guidelines as justification for his definition of authorship. The problem is that academic medicine has set itself apart from the rest of the academic world by using this definition. In the rest of the academic world, when someone makes a substantial contribution to a paper then they are listed on the byline, while the acknowledgment section is typically reserved for people who have not done as much work as the named authors and who did not rise to the level of authorship (e.g., copyeditors, lab technicians). However, in the medical literature, in many cases the roles have been reversed, and those listed in the acknowledgement section have contributed more to the published article than the authors listed on the byline.
In the paper by Wislar et al. , they state that an article is ghostwritten if: “An individual who was not listed as an author made contributions that merited authorship.” This is a very straightforward definition of ghostwriting and implies that merely acknowledging a deserving author in the acknowledgements section is insufficient, and that such papers should be labeled "ghostwritten." Supporting this point of view, Peter Gøtzsche (2009) argued in PLoS Medicine that such acknowledgments are "a euphenism for....XX from Company YY wrote the paper." .
In contrast, if we understand Dr. Jacobs correctly, he believes that a paper with a medical writer mentioned in the acknowledgement section does not constitute a ghostwritten paper. However, this is not universally accepted, and significant ambiguity remains. Is it possible that in the Wislar survey , some of the respondents felt that listing a deserving author in the acknowledgment section was appropriate?
A clarification by ICMJE about the appropriateness of listing deserving authors in the acknowledgment section would be greatly appreciated for the authors of medical papers. It might also be appropriate for the editors of the BMJ, JAMA, or other high-impact medical journals to clarify this simple and straightforward issue.
Jeffrey R. Lacasse
1. Jacobs, A. Re: Honorary and ghost authorship in high impact biomedical journals: a cross sectional survey. BMJ Rapid Response, Monday 2011-11-28.
2. Wislar JS, Flanagin A, Fontanarosa PB, DeAngelis CD. Honorary and ghost authorship in high impact biomedical journals: a cross sectional survey. BMJ 2011; 343:d6128
3. Gøtzsche PC, Kassirer JP, Woolley KL, Wager E, Jacobs A, et al. 2009 What Should Be Done To Tackle Ghostwriting in the Medical Literature? PLoS Med 6(2): e1000023. doi:10.1371/journal.pmed.1000023
Competing interests: No competing interests
Based on this paper, I just posted an opinion on honorary and ghost authorship among Mexican biomedical journals. http://doc2doc.bmj.com/blogs/newsblog/_honorary-ghost-authorship-among-m...
Hopefully, everyday more journals will delineate procedures to strengthen authorship policies and prevent inappropriate authorship from happening.
Competing interests: No competing interests
Jonathan Leo cites a recent blogpost that I wrote in support of the following statements:
"Several individuals and organizations have used their compliance with ICMJE guidelines as a defense against charges that their papers involved ghostwriters. They argue that papers with unnamed authors should not be considered ghostwritten if they otherwise meet ICMJE guidelines."
I don't think Leo can have read my blogpost very carefully, as he has substantially misunderstood what I was saying. I did not say in that article that it is OK for papers to have unnamed authors. As far as I know, no-one else is making that argument either. What I said was that if someone who makes substantial contributions to a paper does not qualify for authorship according to the ICMJE criteria, then it is important that that person be named in the paper. The most appropriate place to do this is usually an acknowledgements section. Failure to be named in this way is indeed ghostwriting, and I do not attempt to defend it.
I realise that Leo does not like the ICMJE criteria and believes that the definition of authorship should be drawn more widely. I am not entirely unsympathetic to that position, but the fact is that at the moment, most journals use the ICMJE criteria to determine authorship, so those are the rules that we must follow until such time as the community of journal editors decides differently.
I encourage others to read my blogpost in full, where I explain these things in more detail.
Competing interests: My company provides professional medical writing services.
The danger with the Wislar paper is that people will start misquoting the results for "ghostauthoring" (7.9%; 49 / 622 papers) as results for "ghostwriting". Has anyone else read the last table on the last page of the supplementary material? Here you will find that the actual prevalence of "true" ghostwriting was only 0.16% or 1 out of 622 papers. There seems to be a minor error in the table (0.3% reported, not 0.16% ie, 1 / 622). Ghostwriting is a problem, but it is a shame that this important datapoint was relegated from the full paper to the supplementary material. One could almost miss it.
Professor Karen Woolley PhD CMPP
Competing interests: I conduct and publish research on ethical medical writing practices. I am actively involved in not-for-profit associations that educate members on ethical publication practices. I am paid to provide ethical medical writing training courses and services for not-for-profit and for-profit clients.
Mark Sillender suggests movie-style credits. It's such a good idea that the BMJ and a few other journals already have them, in the form of contributorship statements:
But it seems they're still not enough. We'll soon be updating our advice on contributorship to prompt clearer declarations about these people's roles in each study: principal investigator(s), other investigators including statisticians, corresponding authors, people who collected data, writers, and other contributors.
Competing interests: I write and update the BMJ's advice to authors
Movie-style credits could solve this long-standing problem.
The research group lead would be the producer and the principal investigator would be the director.
Ghosts could be credited as scriptwriters. Patients would be cast. The helpful medical student could be "best boy/girl", and the Prof would, of course, be the "gaffer".
Competing interests: No competing interests
Posted on behalf of the ISMPP Board of Trustees:
Transparency of authorship is one of the key areas that the International Society for Medical Publication Professionals (ISMPP), a global, non-profit organization that advocates for and educates on best practices, professional ethics, and transparency in medical publishing, is keen to clarify.
ISMPP supports the position that when byline authors have not directed publication development, or when contributors have not been properly acknowledged or appropriately listed as authors, this is inappropriate authorship. However, as medical publication professionals, we regularly encounter investigators and authors who are unaware of the current requirements for authorship and disclosure of potential conflicts of interest.
While the revised good publication practice guidelines (GPP2)1 were developed specifically to guide publication of pharmaceutical industry-sponsored research, they also have relevance and can be applied to non-industry-funded research, and to the honorary and ghost authorship issues raised in this article. We use GPP2 as a tool to define roles and responsibilities at the beginning of the manuscript development process. As medical publication professionals, we are in a position to remind contributors of their authorship responsibilities as we facilitate manuscript development, and can bring it to the attention of the corresponding author if we believe individuals are failing to contribute sufficiently to merit authorship, whether the criteria are ICMJE or journal-specific.
Who would be reasonably expected to fulfil a similar role in an academic setting? If it falls to the corresponding author, then the current article shows that in 2008, 1 in every 5 permitted inappropriate authorship to occur.
The article calls on journals to be more proactive in the matter of inappropriate authorship. Lack of consistency of authorship criteria makes it difficult for authors to realise they may be acting improperly – surely wouldn't adopting a consistent model for authorship, contributorship and acknowledgement be the first step in this process?
1. Graf C, Battisti WP, Bridges D, et al. for the International Society of Medical Publication Professionals. Good publication practice for communicating company sponsored medical research: the GPP2 guidelines. BMJ 2009;339:b4330.
Competing interests: I am employed by the International Society for Medical Publication Professionals, Inc.