Data Money
Data money by Ownipics via Flickr

Medical ghostwriting is difficult to investigate, for the actors involved in conducting, analyzing, writing, marketing and managing a piece of research are unacknowledged in the final manuscript.

What little is known about ghostwriting is based on anonymous surveys, interviews, and the occasional deposition of a disgruntled whistle-blower.  In the event of a lawsuit, when companies are forced to reveal internal documents, we may view the underbelly of this spectral industry, which is thought to be involved behind much of the influential medical literature.

Sergio Sismondo, a philosopher and sociologist at Queens University in Canada, used a different approach to understand medical ghostwriting. Given the propensity for industries to organize, Sismondo simply attended one of their conferences.

In his 2009 article, Ghosts in the Machine: Publication Planning in the Medical Sciences,” Sismondo documented an annual meeting of the International Society for Medical Publication Professionals (ISMPP) to explore the workings and relationships among different interest groups.  The result revealed a complex network of actors including industry sponsors, contract research organizations (CROs), medical communication companies, STM publishers, journal editors, and academics.

And they don’t always work together harmoniously, Sismondo reported.  Industry scientists are often antagonistic toward writers and marketers for misreporting their results to create spin. One presenter — an STM publisher — believed that its own journal editors, and the authorship policies they created, were ultimately doing harm to medical publishing:

We spend a lot of time trying to re-educate our journal editors. . . . We’re saying you have to change your instructions for authors. You have to reflect the changing mood of the times. And yet we still get journal editors who say ‘This journal frowns on ghostwriting’ or something similar. ‘This journal will not accept papers that have writing support.’ And actually what we’re trying to say to them is, ‘Fine, you may have that view but what you’re actually doing is driving it underground. It’s far better to be transparent and get this out into the open.

This is not a picture of a finely tuned machine, but an industry composed of various parts with different values, goals, and agendas.  Even within specific groups, there may be some ambivalence toward ghostwriting.  For instance, journal editors clearly benefit from clear and concise manuscripts crafted by professional writers but also demand adherence to strict authorship criteria, which precluded their listing as authors.

As for academic authors, publication planners view key opinion leaders (KOLs) as little more than indispensable louses:

Although they are recognized as crucial, KOL authors are often portrayed as lazy and greedy. As depicted by planners, they typically make few substantial contributions to the manuscripts they author, are slow to respond, and miss deadlines. They expect prominence in authorship order, and sometimes demand money for their contribution.

To their defense, Sismondo writes, these academic authors are typically provided with only a penultimate version of the article shortly before it is whisked off for journal submission. From the perspective of a publications planner, an academic is viewed more of as a check box on a prepared list of required steps than a true intellectual collaborator.

Sismondo avoids the familiar narrative of painting the issues too simply, as a crisis in scholarly communication, with its heroes and villains and simple solutions.  Nor does he see everyone on equal moral grounds, pointing out serious conflicts of interest, such as when a CRO or academic publisher owns and operates a professional medical communications company.

Stepping back and reflecting on the medical publishing industry, Sismondo is philosophical.

Authorship in the medical sciences has lost its romantic notion of the auteur and has returned us to an earlier version of the scientific author, as the one who speaks with authority.  This is the age of Robert Boyle and his air pump, where scores of assistants and technicians performed the necessary functions of science and yet remained hidden from the scholarly record.

And yet there is one difference.  The academic “author” has been marginalized in the process, sought only for the authority of his name and no longer for his knowledge or expertise.  When operated in this way, medical science is reduced to a form of marketing, which draws its authority from traditional academic science.

Given the investments and payoffs from this new model of corporate science, don’t expect change anytime soon.  For the applied biomedical sciences, this has become the new normal.

Enhanced by Zemanta
Phil Davis

Phil Davis

Phil Davis is a publishing consultant specializing in the statistical analysis of citation, readership, publication and survey data. He has a Ph.D. in science communication from Cornell University (2010), extensive experience as a science librarian (1995-2006) and was trained as a life scientist.


5 Thoughts on "Ghosts in the Machine — The Industry of Medical Authorship"

This article paints a picture that doesn’t match the reality very well. Let’s take a couple of examples.

“academic authors are typically provided with only a penultimate version of the article shortly before it is whisked off for journal submission”

That’s a popular stereotype, but not how it operates in my experience. We always have input from academic authors right from the stage of writing an outline, before the first draft of the manuscript is written. I can’t speak for the whole industry, of course, but I believe our experience to be pretty typical of how most companies operate these days.

“This is not a picture of a finely tuned machine, but an industry composed of various parts with different values, goals, and agendas. Even within specific groups, there may be some ambivalence toward ghostwriting.”

There may well be different people with different agendas in the pharmaceutical industry, but when it comes to ghostwriting, there is actually an astonishing level of consensus. Everyone agrees that ghostwriting is a bad thing. That’s not just in the industry, that’s pretty much everyone else as well. Just read any of the many guidelines that mention ghostwriting (EMWA, GPP, PhRMA, ICMJE etc) and try to find one that differs significantly from any of the others. You won’t.

I’m not a industry analyst, so I will cite from Sismondo’s paper (it is freely-available). “BJ” is listed as a “an efficiency expert talking about metrics” at the conference:

According to speaker BJ’s estimate, 50% of companies show only the penultimate manuscript to authors, to solicit their input.(p.187)

Your second point, about ghostwriting being generally viewed as “a bad thing” is well-taken, but industry standards and compliance are two different issues. The guidelines you cite have no power to enforce their rules or sanction those who break them.

In spite of all of these well-intentioned steps, ghostwriting appears to be a well-documented fact:

Flanagin, A. et al. 1998. Prevalence of Articles With Honorary Authors and Ghost Authors in Peer-Reviewed Medical Journals. JAMA 280: 222-224.

Gøtzsche, P. C., Hróbjartsson, A., Johansen, H. K., Haahr, M. T., Altman, D. G., & Chan, A.-W. 2007. Ghost Authorship in Industry-Initiated Randomised Trials. PLoS Med 4: e19.

Mowatt, G. et al. 2002. Prevalence of Honorary and Ghost Authorship in Cochrane Reviews. JAMA 287: 2769-2771.

Ross, J.S. et al. 2008. Guest Authorship and Ghostwriting in Publications Related to Rofecoxib: A Case Study of Industry Documents From Rofecoxib Litigation. JAMA 299: 1800-1812.

OK, on the first point, about 50% of companies showing only the penultimate manuscript to authors, there is absolutely no information about where that figure came from. As a statistician, I do not trust figures unless I know the methods behind them. Without further information, I shall treat that figure of 50% as being something that was made up on the back of a fag packet (which, in fact, it may well have been).

On the second point, you are right that standards and compliance are not the same thing. It’s true that, although everyone agrees that ghostwriting is bad, it still happens. We don’t really know why it happens. Possible explanations include people either being unaware that the guidelines exist, or deliberately ignoring the guidelines. And if it’s deliberate, then the concealment could be on the part of the writer, the sponsor, the named authors, or the journal (experience suggests that the named authors are the prime suspects there, although I have no reliable data to back that up). My own research shows that lack of awareness of guidelines is certainly part of the problem, as writers who are unaware of guidelines are less likely to be acknowledged than writers who are familiar with guidelines. However, I’m sure it’s more complicated than that, as even writers who are aware of guidelines are not always acknowledged. I do actually have some qualitative data that might help to investigate the reasons why writers are not always acknowledged, but have yet to analyse it. If you know of any other research on this, I’d be very interested to see it.

The other interesting question, of course, is how common ghostwriting is. Unfortunately, the papers you cite don’t really give much insight. Flanagin et al was a good study, but is now hopelessly out of date. The landscape has changed dramatically since 1998. Gøtzsche et al was not about medical writers, it was about ghost authorship by statisticians. That’s a whole other problem, of course, but not really relevant to the present discussion. Mowatt et al is also out of date, and of course Cochrane reviews are very different to the wider population of medical literature. Ross et al had no quantitative data in it: it was simply an in-depth description of a small number of highly non-randomly-selected examples.

My own research, based on a survey done in 2008 (see for details), gives a tentative estimate of the proportion of contributions from medical writers that are unacknowledged (ie ghostwriting) as 42%, although that’s probably an underestimate because of some unavoidable biases in the survey. Of note, that’s a substantially lower number than the proportion in 2005, which suggests that ghostwriting is becoming less common.

Note also that that is a figure for the proportion of papers with contributions by professional medical writers that are ghostwritten. It provides no data on what proportion of all papers are ghostwritten.

Clearly, more needs to be done to make sure the guidelines are better observed. I and some colleagues have made a practical suggestion in that direction, although sadly, there seems to be little appetite among medical journal editors to act on it.

See for details.

I just finished reading your article surveying members of the two medical writers associations and feel that this study deserves more recognition. It does indeed fill a missing piece in the literature. It is unfortunate that it was published in your association’s magazine and not in an academic journal. Had I known about it, I would have cited it. Thank you for this contribution.

The main results of your study corroborate with data presented — ghostwriting is still a widespread practice. Your comparison of 2008 with 2005 suggests that unacknowledged writing is declining for association members and propose that adherence to guidelines — especially for writers who work for medical communication companies — may be the reason for the decline.

In sum, we appear to be arguing about related points: we both agree that ghostwriting (and ghost authorship) exists and persists, although you provide data that this practice may be in decline. I look forward to a follow-up survey.

Comments are closed.