On ghostwriting

A long time ago, when the world was young, I did a research project towards my Biochemistry Part II exam. Three months in a lab, learning how to do cell culture, how not to kill myself with iodine-125, and making hundreds of very pretty immunofluorescence photographs. It was a nice little project, and I got some publishable results. So publishable in fact, that when the head of department said he’d be able to find me some money to come and work in the lab over summer—after my exams and before I started my DPhil—I leapt at the chance.

That story, and of how he didn’t pay me for that summer, and how he muscled in on the author list and conspired with my supervisor to push me off the coveted first author position, are for another day. What concerns me now is that I didn’t write the paper. Naturally—I was studying for exams, about to go and work for someone else, and thought it entirely natural that my supervisor (with so much more paper-writing experience than me) should write up my results. Of course I read it, of course I commented—I may even have written up some of the methods—but there’s no way I could be said to have written that paper.

Similarly, even when I was the lead author, having done the intellectual lifting and most of the lab work, I didn’t write my next couple of papers (it could be argued that I was under the anally retentive thumb of certain lab heads, but I couldn’t possibly comment). In fact, my name has appeared on a couple of papers where I wasn’t even in the same lab.

Shocking?

Well, Martin Raff once described the way I was screwed over by that head of department and supervisor as ‘shocking’, but the fact that I didn’t write papers with my name on them isn’t abnormal, is it? In fact, I’ve seen papers in PubMed where it would have been impossible for all the authors to say they had ‘written’ the paper. You know the ones I mean: generally genetics or physics papers with over a hundred co-authors. In those cases it is highly unlikely that the person who actually wrote the paper did any lab work at all (and their intellectual contribution might not have been that great, either).

Equally, I have contributed to many papers where I have not been listed as an author. I have helped foreigners with their English, made comments on style and grammar; even suggested extra figures or experiments. Sometimes I’ve been acknowledged, sometimes I haven’t.

So why does everybody get upset when papers appear that are ‘written’ by someone who isn’t listed as an author?

Dr Vallance, of Ninewells Hospital, Dundee, wrote to Eye (published by NPG, by the way), saying that publication of a particular article (Preclinical aspects of anti-VEGF agents for the treatment of wet AMD: ranibizumab and bevacizumab) is “a matter for the Editor.” Why? Of the review article, Dr Vallance says, “My main concern is that the authors did not write it.” He adds, “The subject of ‘medical ghostwriting’ and the potential effects on healthcare delivery is a controversial area.”

There is a robust response from the credited authors of the article in question, accusing Dr Vallance of libel. There is an equally robust response from the medcomms agency that was tasked with the actual writing of the article.

Medical communications and publishing, and the pharmaceutical industry itself, are among the most heavily regulated industries on the planet. And writing of articles by people who do not appear in the authorship list goes on all the time—in an orderly and heavily regulated fashion.

Pharmaceutical companies fund a great deal of research, and they publish that research in peer-reviewed journals. They also publish the results of clinical trials (admittedly there is a problem with which results get published, but that’s a separate argument). When a study, pre-clinical or otherwise (or, as in this case, a review), is published as a peer-reviewed article then the financial involvement of that pharmaceutical company is disclosed—as are the financial interests of any and all of the authors under whose names the article appears. When anybody performs research, and wishes to publish the results of that research, why should they not retain the services of professional writers? Especially when they have big pharma to pay for those services? Each author takes responsibility for their contribution to the research and the manuscript as a whole (and please, do read the response by Meyer and Holz).

If the person who actually typed the words into Word in the first place happens to be a professional writer, and this contribution is acknowledged, then why is there a problem? Wouldn’t you rather see data published quickly, knowing that the brains behind the research are continuing to contribute to advancing healthcare? Or would you rather wade through execrable prose? Why is this any different from helping a colleague edit their manuscript? (Except that money changes hands. In which case there is a paper trail.)

This isn’t making up results, this isn’t cherry-picking data, this isn’t creating fake medical journals; this certainly isn’t writing about a drug and then putting top docs’ names to it.

Let’s turn this around slightly. If you, as an author, enlisted a colleague’s assistance in editing or writing your manuscript, and remunerated them in some way, would you put them as an author? Wouldn’t you rather acknowledge their help at the bottom of the manuscript? And you would even dare say they were, or how they were, remunerated?

Not to mention that if this practice was banned, all these highly skilled writers would be out of a job, and instead likely competing for the same grants you’re after.


Conflict of interest: rpg works in the med comms & publishing industry.

About rpg

Scientist, poet, gadfly
This entry was posted in Rants and tagged , . Bookmark the permalink.

11 Responses to On ghostwriting

  1. cromercrox says:

    I’ve heard of mixed metaphors, but ‘anally retentive thumb’ has to be the worst.

  2. Austin says:

    I ghost-wrote an obituary once. Strange but true.

  3. Nice commentary Richard.

    The answer to your last query – no, I wouldn’t list them as an author, I’d acknowledge them in the appropriate place (the, er, acknowledgments section). Same as all of the people who acknowledge core facility services that we provide – we sequence some DNA for you on a cost-recovery basis, and provide the data to you. You pay us for our time and materials. You own the data, you interpret it, draw conclusions, write it up, publish it, and maybe even own any arising intellectual property. We provided a service, for a fee, that’s it. No authorship position. I see fee-for-service writing as the same.

    Some institutions have people who are paid, possibly among other duties, as grant-writers. That in my mind isn’t really different from a paid manuscript writer, either, although I suspect that in the latter case authorship would often be given as a courtesy. I’d like to hear Cath’s take on all of this.

  4. rpg says:

    Actually Winty, sequencing was one o the other non-author examples I meant to give. Thanks.

  5. Nice article. I can’t imagine having a pro- writer write up my results, but then again this isn’t at all normal in the places I publish. If I pay a company for say analytical analysis the in a paper, no I wouldn’t list them as an author. If a colleague edits my paper yes I would. But it also depends on what they want. The important thing is you discuss it with your collaborators. I for instance did alot of experimental work for someone once and when it came to paper time I insisted they took my name off the paper, because I had no clue about, or the time to read their analysis of the results. I didn’t want to be on a paper I didn’t know much about. I think personally if you put someone’s name on a paper, you must always tell them before you submit it…

  6. Nico says:

    Just yesterday I talked with a friend who has solid, interesting results, but between the fact that he is swamped by admin (course leader…) and that he is under little pressure to get grants (unusual, he does relatively low-cost research), he has a pile of stuff to be written but doesn’t get round to it. I would see paying someone to help with the write-up as a reasonable way out, you can’t rely on slaves minions students/postdocs to do it all the time.
    After all it surely is better that the data is out there, rather than locked in his hard drive?

  7. Nico, you’ve made me think of another angle to all of this. Much as we love our postdocs, it is true that written English is not the strong suit of some of them. I could imagine a lab where it might be more effective to retain a third-party writer for at least some of the writing, rather than fight through multiple edits from a non-native English speaker.

    Although, that said, part of postdoctoral training should be in learning to write papers and grants in the vernacular, so perhaps my argument here is a poor one after all.

  8. Nico says:

    Richard (W), I actually think that writing papers should be part of doctoral training, if only because it is good training for the thesis! However, there are many people for whom it takes a lot of effort to write well, for example because of dyslexia (my dad struggles with English even after 35+ years in research, one year in the US and English-speaking daughter in law!), or more simply because their first language isn’t scholarly English. I have no problem with giving these people a little help, whether they hired it upfront to help writing the paper or afterwards to clean up the English (usual plug to Nature Language Editing, other editing companies are available etc). And once the peer-review hurdle is cleared your friendly copy-editors will do their best as well 😉

  9. Nico – agreed, doctoral training.

  10. Cath@VWXYNot? says:

    Very thought-provoking post, Richard! This is an interesting and complicated subject.

    My own feeling is that there’s nothing wrong with using a professional writer to help prepare manuscripts, in the private or public sector (I would say that, given that it’s part of my job description!). However, I feel that if the hired gun was the primary author – if they were given just the figures and other bare bones of the methods etc., and wrote up the whole thing basically from scratch – that should be disclosed somewhere, probably in the acknowledgments section. However, if they (like me) were more of an editor who worked on a draft – even a rough draft – written by a member of the study team, then I’d say a full disclosure of the type of contribution isn’t necessary, although ideally the writer would be thanked in the acknowledgments section.

    I have a strong preference for the latter situation, by the way. I think a paper should ideally be drafted by someone who was closely involved in the conduct of the research – preferably a trainee under an appropriate degree of supervision depending on ability and experience, or their supervisor if necessary. As someone who assists with many papers per year across a department of five PIs with extremely diverse interests – everything from pathology to genomics to statistics to algorithm development to radiochemistry and functional imaging studies that require me to know about cyclotrons and annihilation photons – I know that I don’t have the depth of expertise in any given area to draft a paper from scratch. Someone who’s much more focused than I am needs to do that job. I prefer being a broadly-based generalist with some expertise in multiple areas – it just suits my personality and interests better – but that makes me better suited to editing tasks and writing lay summaries etc. than to de novo generation of the complex, technical text needed for papers and grant proposals.

    My own input varies massively from paper to paper – everything from major edits to a simple proof-read – and sometimes I’m acknowledged, but at other times I’m not. I don’t worry too much about it, although I do appreciate it when mentioned! The two instances where I’m an actual author (both accepted recently and now in press!) involved 1) making a figure and a table based on the statistician’s results and drafting all but one section of the manuscript on a project where I’d also worked extensively on the grant that funded the study, and had influenced its design in the process; and 2) a review paper (where the bar for authorship is lower) which I edited extensively, proofed, and suggested two new sections.

Comments are closed.