ghostwriter ~ 6
Ghostwriter by striatic via Flickr

I’ve met a ghost.  Not the kind who haunt old Victorian mansions, moaning, pacing hallways, slamming doors, but a family man with an ordinary day job.

He works as medical writer, and in the last seven years, he has authored an impressive set of articles found in top scientific journals. You won’t recognize his name, because he is unacknowledged for his contributions.

He is known as a “ghostwriter,” a figure despised by medical journal editors who wish to see him “exorcised” from the literature.

Because they leave few traces, little is known about ghostwriters.  Luckily, I had the privilege of interviewing one who agreed to speak freely about ghostwriting on condition of anonymity. For All Hallows’ Eve, we present you with an Interview with a Ghost.

Q: Can you briefly describe what you do as a medical ghostwriter?

A: I work as part of a publishing team consisting of: 1) the supporting pharmaceutical company, 2) the true authors of the study, 3) reference support staff, and 4) graphic support staff, if I’m lucky. In a perfect world, this team would work together to produce a readable, accurate account of the study in question, its results, and a critical review of how those results may (or may not) change the standard of care. In reality, authors are required to publish their findings, and the pharmaceutical company is paying for the study, so they are looking to me to produce a manuscript that will help make the company’s drug the new top dog.

Q: What led you to medical writing?

A: When I was a graduate student, I had no idea there was anything else for a PhD in biochemistry and molecular biology to do other than an industry lab, a government lab, or staying in academia. My first position, a publications manager for a medical communications company, just sort of fell in my lap.

Q: What do medical writers charge for their work?

A: Experienced medical writers will typically charge between $100 and $150/hr or about $3,000 to $4,000 per assignment.

Q: In cases where you are not listed as an author, are you listed as a contributor in the acknowledgments section of a journal?

A: Some clients feel it is in their best interest to present me as an author, but the pharmaceutical company will often shut them down, since I am not a thought-leader in that field. Sometimes I am added to the acknowledgments; most often I am not mentioned at all, which technically makes me a ghost.

Q: In what journals have articles that you’ve ghostwritten appeared?

A: My work appears in almost every medical field, from the New England Journal of Medicine, to the Journal of Clinical Oncology, Oncology, PLoS Medicine, American Journal of Psychiatry, Cardiology, and supplements to most of these journals as well. I have also authored a number of online continuing medical education (CME) sessions, various symposia presented at medical conferences, as well as posters, abstracts, and those pamphlets you see at your doctor’s office discussing treatment for a disease or how to self-medicate.

Q: Are there any tell-tale signs of a ghostwritten article? Can you detect it in other papers?

A: Ghostwriting is easy to detect if you know the primary author, or witnessed that individual present at a conference. When someone isn’t clear, or jumps around a lot, or can’t speak the language, yet has a publication in the NEJM, you can bet there was a ghost involved. Very prolific clinical researchers – especially those actively seeing patients and running their own clinics – are also likely to be working with ghosts.

Q: Studies of the prevalence of ghost and honorary authorship in medical journals [1], and Cochrane reviews [2] report that about one in 10 articles include ghost authors. Both of these studies were based on author self-reports. How accurate do you think these estimates are?

A: I think that figure is a little low. You have to remember that authorship is key to success and survival for most researchers.

Q: Litigation against Merck regarding Vioxx revealed that academics were routinely recruited and paid to put their names on articles they had little (if any) involvement in producing, for the explicit purpose of creating credibility for the study [3].  Is this a common practice?

In my experience, the pharmaceutical company would pay a communications/marketing company to write the manuscript, who would then go out and find academics who would be willing to become the “authors” of the manuscript and paid an honorarium. I’ve worked with some authors who do absolutely nothing on the manuscript, requiring an additional ghostwriter to be hired, and still demand an honorarium for their time. These academics are willing to enter into this relationship because of the importance of authorship to their careers. You can’t entirely blame the pharma company. Universities encourage academics to play this game.

Q: Do you believe that you add value to medical publishing, or are you merely a hired hand for work that others refuse to do or do badly?

A: Short answer is “yes.” I believe I provide a service to those who need assistance presenting their findings to the scientific community. If you have a great study but present it badly you won’t be seeing it anytime soon in NEJM. Until we do away with authorship requirements for promotion and tenure all together, we will continue to need the skills of individuals, like myself, if these studies are to be published at all.

Q: Have you attempted to influence an editor to accept a manuscript?

A: I personally have not. Besides being morally wrong in my view, it is extremely difficult to get in touch with journal editors, which makes the submission process such a joy. Yes, that was sarcasm.

Q: Have there been cases where your sponsor wanted you to make conclusions that were not supported by the data? How did you deal with these instances?

A: Yes, I have. Every manuscript includes some marketing spin and controlling how much spin is the trick. Sometimes the abstract — often written by someone else — comes to a different conclusion than the main manuscript or relies on different data. The easiest way to deal with the issue [of unsupported conclusions] is to use the data.

Q: Are you a member of a professional society for science writers? Does that society have any guidelines for ghostwriting?

A: I have been a member of the American Medical Writers Association from time to time. They do have a series of guidelines and a code of ethics and they do suggest that medical communicators be recognized for their contributions.

Q: How do you claim credit for articles that you have ghostwritten for which you are not acknowledged? What does your CV or resume look like?

A: My CV consists of the various positions I’ve held through the years, and I do have a portfolio of projects that I have worked on. Most of my new business comes from former clients. If you stop hearing from a company that you have a long working relationship with, it may be due to some issue they had with your last project.

Q: Do you trust the medical literature? Or, now that you have a view on how the system really works, does it give you pause when you read the literature?

A: Every manuscript uses some marketing gimmick or spin in order to enhance the weight of the data. I remember when I was a graduate student, we were taught to look at our data with a critical eye. When a figure from a scientific lab has extra bands or markings that are not even mentioned in the text, it throws the whole manuscript in question for me. I can’t stop questioning the literature, especially in the higher tier journals, where there is a lot of “I’ll wash your back if you wash mine,” mentality. Don’t get me wrong, I love science, I just know how the game is played, and it leaves a bad taste in my mouth.

Q: Some have proposed that we need to rid the medical literature of ghostwriters; others argue that we merely need to make their identity and contributions known [4]. What is your opinion?

A: Either journals should establish agreed-upon rules for what defines an “author” or do away with authorship altogether. I believe that everyone who worked on the paper should be mentioned in the Contributors section.

Q: Can you see your reflection in a mirror?

A: Well, since I’m nocturnal, there is never enough light for me to see anything in a mirror. I can walk through walls, though.

Q: What do you think of garlic?

A: I love it.  Remember, I’m not a vampire.

Q: What are you dressing up for Halloween?

A: A medical researcher. Or the fifth dentist (4 out of 5 dentists agree . . . ).

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.


30 Thoughts on "Interview With a Ghost (Writer)"

I wonder how copyright is handled? Does the pharma company commission the work of the ghostwriter and designate it a “work made for hire,” in which case the company becomes the legal “author.” But then how is this legal fact acknowledged in the article itself? The listed “authors” would then not, legally, be the real authors, including the ghostwriter.

If the matter is instead handled as a transfer of copyright from the ghostwriter to the named authors, this fact would have to be acknowledged in the form used by the U.S. Copyright Office for registering the copyright. That would, presumably, allow for a determination of how many such articles are ghostwritten, if anybody cared to undertake the research. Now that registrations are done online mostly and the records open to public online inspection, this wouldn’t be too difficult to undertake.

From this description, ghostwriting would appear to be just a more extreme case of regular copyediting, which sometimes can involve substantial rewriting, if not writing de novo.

I do hope people reading this don’t get the impression that this is how all medical writers work. Being unacknowledged and being complicit in “spinning” results is very much frowned on by the mainstream medical writing profession.

Reference 4 in the article is a pretty good one for what the medical writing profession is doing to try to make sure that people like the ghost in this article clean up their act (declaration of interest: I was one of the authors of that paper).

I spent a short time as a medical writer and the company I worked for was very hot on working on Good Publication Practice. Having said that, I was still unhappy that I would often have to provide intellectual input into a publication whilst many of the ‘official’ authors did little more than nod in approval. Medical Writers should be full authors but what the Pharma companies do is in effect ‘hire’ big names medics as authors, for their prestige.

Personal responsibility. It is up to every professional medical writer, and other stakeholders with a vested interest, including the interviewee in this article, to fix the impression that all editorial assistance is “ghostwriting.” It is not. There are numerous guidelines (ICMJE, GPP2, etc)that define who should be included as an author and how editorial assistance should be acknowledged. I would like to see an intereview with one of the thousands of professional medical writers who consider personal ethics an imporant part of what they do.

I wouldn’t go as far as to call the interviewee “unethical.” Indeed, it appears that the decision to include writer’s name on the final paper is outside the writer’s control. The following response indicates this point clearly:

A: Some clients feel it is in their best interest to present me as an author, but the pharmaceutical company will often shut them down, since I am not a thought-leader in that field. Sometimes I am added to the acknowledgments; most often I am not mentioned at all, which technically makes me a ghost.

Has your name been included on every single paper to which you have contributed? If not, I’m not sure how your “personal ethics” will get you back on the author list. The guidelines are just that — guidelines. They have absolutely no weight to enforce standards.

Come on Phil. The interviewee had a choice. If the client would not allow appropriate acknowledgment, then the writer could turn down the business – that is ethics. This should be established right up front and if the client does not agree, then the writer walks.

I find it very interesting that an article implicitly concerned with transparency is formatted as an interview with an anonymous “ghostwriter”. Many of the practices described by this “source” are rather out-of-date, as most major pharmaceutical companies now go to great pains to ensure medical writing contributions are acknowledged (and rightly so). I am less than convinced that this article represents anything except the interviewer’s own incomplete knowledge, misperceptions, and desire to “stir the pot”.

While you should be thanked for your lip service to the pharmaceutical industry, the results of this interview are anything but “out-of-date.” Consider the results of this study:

Of 44 industry-initiated trials, there was evidence of ghost authorship in 33, increasing to 40 when a person qualifying for authorship was acknowledged rather than appearing as an author.

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.

That paper, despite the title, had very little to do with medical writing. It was mainly about whether statisticians were listed as authors. The interview was with a medical writer, not a statistician.

Phil, your insulting comment aside, you say this is “anything but out-of-date”, but the very source you cite provides the dates of the studies it examined as 1994-1995. Do you seriously consider studies from 15 years ago as contemporary?

Fair enough, Ed.
Here is a 2008 study [1] that focuses on the lawsuit against Merck and relying on documents between 1996 and 2004. The two health communications companies named in the report are still in business.

Now, it may be that the industry has changed but just not reflected in the literature. So the burden of proof now rests with you to demonstrate that your industry has indeed changed.

[1] 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.

I don’t think the Ross et al paper is very relevant. Apart from the fact that it’s also out-of-date (although published in 2008, the papers it looked at go back to 1996), it has no quantitative data. It is simply a descriptive study of some papers that were selected on the basis that they were involved in a lawsuit (and therefore extremely unlikely to be a representative sample).

Response by Phil Davis (as we have run out of discussion thread):

I completely follow that the studies I’m citing may be out of date (although, 1996 was just the starting point of the research — 2004 was the end-point), and that these studies may not be statistically representative of all medical authorship. However, we are attempting to document a publicly opaque relationship and are dependent, in large part, upon litigation to provide us with public evidence. And yet, surveys and interviews still acknowledge that ghost authorship is an ongoing problem. Unless you and your industry can provide evidence that the practice has ended, we need to assume that things haven’t changed. Put another way, “out-of-date” is not a sufficient counter-argument.

As correctly pointed out by Adam, this is not evidence that the practices are still widespread. Phil, since you wrote this article implying that the practices are still an industry standard, the burden lies with you to provide the evidence that your “ghost” has any substance. It may be telling that, despite the current level of public and press scrutiny about this topic, you seem to be experiencing difficulty in providing any evidence to that effect.

Response by Phil (since we have run out of commenting thread):

While I provide evidence in the form of surveys, interviews, bibliometric studies, internal documents, and conference proceedings, you are only able to offer incredulity. If we’re going to continue the discussion, you’ll need to bring more to the table than blind support for Adam.

Phil – have a read of the following article to get a feel for the other side of the story. I know the author and know he had tremendous difficulty finding anyone brave enough to publish this. Even the lay press, who jumped all over the JAMA story, refused to allow so much as a repsonse. Why? In my opinion, it is because it contradicts much of what they have written and would expose their shoddy journalism and because it does not sell papers.

Can a ghostwriter use breaking information about a drug to buy stock in advance and stuff like that or are there agreements to that effect?…just curious…

Sameer, that would constitute insider information and is a violation of the law, not just “agreements”.

I agree with other replies that this individual must be one of the very very few remaining old-school ghosts that brings what is now a transparent and acceptable service into disrepute. I wonder if this article makes the editors of the named journals as furious as me – how dare this individual disregard industry-accepted standards and principles of the ICMJE, Good Publication Practice, International Society of Medical Publishing Professions (ISMPP) and others and talk as though they represents the majority. To say that most universities encourage academics to use ghost-writers when many explicitly forbid it is simply inaccurate.

This individual is partly uninformed and wholey well behind the times. I also know for certain this un-named person doesn’t work for my company nor for any of my clients. My advice: time to retire, step-back into the shadows and let the ethical and compliant of us continue to help authors and industry ensure product data is communicated in a timely and unbiased manner.

Healthcare professionals and patients need to rely on what they read in medical journals as to be the whole truth and if you do not believe that in that, you shouldn’t be a medical writer.

Thank you for your contribution.
I cannot comment on whether the ghostwriter I interviewed is representative of your industry. Your perception about the unwillingness of academic medical centers to engage in provisional contracts with industry, however, is not supported by data. In a survey of medical school administrators [1] the researchers report:

“There was considerable disagreement about the acceptability of provisions allowing the sponsor to insert its own statistical analyses in manuscripts (24 percent allowed them, 47 percent disallowed them, and 29 percent were not sure whether they should allow them), draft the manuscript (50 percent allowed it, 40 percent disallowed it, and 11 percent were not sure whether they should allow it), and prohibit investigators from sharing data with third parties after the trial is over (41 percent allowed it, 34 percent disallowed it, and 24 percent were not sure whether they should allow it).”

[1] Mello, M. M., Clarridge, B. R., & Studdert, D. M. 2005. Academic Medical Centers’ Standards for Clinical-Trial Agreements with Industry. N Engl J Med 352: 2202-2210.

Phil Davis says: “I cannot comment on whether the ghostwriter I interviewed is representative of your industry”

So maybe, Phil, instead of presenting this one person with the implicit inference it is widespread, or representative of broad practice today, could you be more responsible and ensure fair balance in your reporting? You could include people working to the now well established industry norm, according to Good Publication Practice?

The topic of the piece was ghostwriting (meaning, unacknowledged authors and contributors) and not medical writing and publication planning in general. This latter topic is covered in a follow-up post,see Ghosts in the Machine.

In spite of Good Publication Practice, there is evidence that ghost authorship is still a problem — perhaps a very large problem — in the industry. And while the public has access to individuals like you, societies and associations, and the documents it produces, there is very little known about whether these people and their good intentions are actually changing things for the better.

This interview suggests that you have a long way to go.

Phil, since we ran out of string, I’ve started another. What you’ve offered is an interview with an anonymous source, a outdated reference (as also pointed out by Adam, it is not especially relevant because it mostly identified omission of statisticians from authorship), and internal documents from a lawsuit whose interpretation is not without question (e.g., see reference from Dan). You have provided no credible evidence that “ghostwriting” a current problem, much less a widespread one.
There is little doubt that there were some practices in the past that were questionable at best, although there is some debate as to the extent of the problem. Nevertheless, the adoption of guidelines by journal editors and others have gone a long way toward providing transparency and accountability. From my experience with industry clients, those I work with are careful to the point of paranoia to ensure proper attribution of contributorship. On the other hand, there are still issues that deserve further discussion (e.g., the issue of authorship vs acknowledgement, as alluded to by MT, above). However, this “interview” does little to address or even publicize any current issue, but rather it seems designed simply to stir up mistrust.
I wonder how you would react if, in the wake of the Jayson Blair scandal (faked stories for the NY Times), an “anonymous” interview was posted implying that such fabrication of stories and issues by members of the press was a widespread problem. In fact, it seems unlikely that it was entirely an isolated incident. I also wonder if you would recognize any responsibility of the person that posted that interview for providing evidence as to its credulity, or if you would it acceptable that the interviewer simply adopt the position of “prove me wrong”.

You still haven’t brought anything of discussion to the table with the exception of your outrage, which won’t go far if we are to debate the issue.

Adam Jacobs has alerted me of a survey [1] of members of the European and American Medical Writers Associations (EMWA and AMWA), which fills an important hole in the available information. While Adam reports that the frequency of self-reported ghostwriting is in decline (based on comparing the results of 2008 and 2005), survey respondents still set the mean rate at 42%. I will avoid pointing out the limitations of the study or simply calling it “woefully out of date” for theatrical purposes. For all of its limitations, it is a useful and informative study.

As to the anonymous source, confidentiality was traded for useful insider information, and this practice is not uncommon. Even in Adam Jacob’s study, he provided anonymity to his survey respondents in order to receive credible and honest answers and to avoid self-incrimination. Both forms of gathering information ultimately rely on trust. Do I trust my source? I am an outsider to this industry, but his answers appear to corroborate with information pieced together from other sources. If you wish to believe that all other sources are biased and thus cannot provide absolute “proof” of ghostwriting, this is your prerogative. I’m certainly not going to change your mind.

[1] Jacobs, A., & Hamilton, C. W. 2009. Decreased evidence of ghostwriting in a 2008 vs 2005 survey of medical writers. The Write Stuff 18: 118-123.

Phil, “outrage” is a little strong – it’s more like irritation with having been impugned, along with my colleagues. I also do not object per se to the use of an anonymous source, but rather to the use of that information to imply a widespread problem without sufficient credible and current evidence to back it up. Because you are an outsider, I would have thought that it would be especially important to determine the veracity of your source. As to your position that the onus is on others to prove you wrong, obviously it is often difficult to prove a negative, and I believe that there is a Latin phrase to describe an argument made or position taken due to the lack of evidence of the contrary – argumentum ad ignorantiam. As I implied above, perhaps you should prove to us that journalistic misconduct is not widespread.

Comments are closed.