Thursday, April 17, 2008

Side Effects: Merck is not the first pharmaceutical company to use ghost-writers

An article in The New York Times a few days ago about Vioxx almost makes it sound as if the ghostwriting of new drug studies by pharmaceutical companies is a new and relatively unknown practice. In fact, the ghost-writing of research by pharmaceutical companies eager to publicize positive results about their new drugs in prestigious medical journals has been a widespread industry practice for years.

David Healy, a psychiatrist and professor at the University of Wales in England, was among the first to write about the practice of ghostwriting by drug makers a number of years ago. In several scholarly articles, Dr. Healy drew attention to the fact that Pfizer had ghost-written a series of positive research studies for its blockbuster antidepressant, Zoloft. In a review of internal Pfizer documents unsealed in lawsuits, Healy found that just as Merck did with Vioxx, a Pfizer was preparing a number of Zoloft studies for publication before they had identified authors for the studies. On an internal list of these in-process studies, Pfizer had put question marks next to the line for possible authors.

As my forthcoming book reveals, GlaxoSmithKline resorted to a similar scheme in its effort to publicize the results of a study purporting to show the effectivenes of Paxil in treating depression in adolescents. Unsealed documents from a lawsuit against GlaxoSmithKline show that a consultant for Glaxo ghost-wrote the first draft of this research and then sent it to the principal authors for review and approval. The Paxil study was rejected by JAMA(in large part because its positive conclusion about the antidepressant was not backed up by the actual data in the study), but the study was eventually published in the Journal of the American Academy of Child Psychiatry.

The article in the" based its news on a recent study in JAMA that looked at documents unearthed in lawsuits over Vioxx, which was pulled from the market in 2004. As these documents show, Merck also hired ghostwriters to draft Vioxx research studies, then the company found prominent authors who would be willing to attach their names to the studies for publication in medical journals. As David Egilman, an associate professor in community health at Brown University, tells Pharmalot blogger Ed Silverman, “By having no outside researchers look at data or be in control of a study allowed Merck to manipulate the trials. They were able to use spin to create the impression that the drug didn’t kill people, which it did."

As Dr. Joseph S. Ross, the lead author of the JAMA article says, this practice raises broad questions about the validity of much published drug research. Indeed, I would argue that it raises questions about the credibility of the entire system of drug testing and research. And now, finally, the mainstream media is paying attention.


Rodney Wilson said...

I had no idea there were ghost-written studies. Where are the checks and balances!?

Anonymous said...

Of course, all this depends on how you define "ghosting": if the problem is that lead authors are not doing the bulk of the actual text production, but _did_ run the study, that seems different from a case where the lead authors were not involved with the data collection or the writing, but instead are figureheads for a manuscript prepared before they were ever approached.

But if your problem is really the outsourcing of publication strategy from pharma companies to publication management companies, then your argument should be with ISMPP and the companies affiliated with it.

The baseline problem from my perspective is that peer reviewers at many journals do not have the ability to tell the difference between "pure" review or research articles and those that are really closer to product development reports. That problem is NOT the fault of the writers, no matter what their genuine affiliations.

Michael S. Altus, PhD, ELS said...

A Primer on the Three “G” words: Ghostauthoring, Ghostwriting and Guest Authoring.

First, we need to start with basics, the often-quoted Uniform Requirements for Manuscripts Submitted to Biomedical Journals: Writing and Editing for Biomedical Publication of the International Committee of Medical Journal Editors (

II. Ethical Considerations in the Conduct and Reporting of Research
II.A Authorship and Contributorship
II.A.1. Byline Authors

Authorship credit should be based on 1) substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data; 2) drafting the article or revising it critically for important intellectual content; and 3) final approval of the version to be published. Authors should meet conditions 1, 2, and 3.

Now to the three “G” words. The focus is on concealment and its obverse, disclosure. It depends on what is being concealed (disclosed).

Ghostauthoring occurs when the responsibility for the work of one is attributed to another. A commonly discussed example is when a pharmaceutical company, typically through a medical education and communication company, develops an article (an account of original research or a review of and opinion about research). The names of some who should be identified as authors are concealed.

For an instance of a pure-and-simple instance of ghostauthoring, go a Wall Street Journal Health Blog article, “Odd Ghostwriting Offer Raises Researcher’s Blood Pressure” (; November 21, 2007).

I really with that the WSJ bloggers had used the word “ghostauthoring” instead of “ghostwriting”. It’s a lost cause, though, to get this worthwhile distinction to be accepted. “Ghostwriting” is used when “ghostauthoring” is meant. We have much to be grateful to a highly regarded medical communicator, Tom Lang, for establishing this distinction (The Role Of Professional Writers;

Ghostwriting, per the distinction here, occurs when “a professional medical writer assists in presenting the content of a named author” (per Lang) but is not noted in the acknowledgments. Lang is the only authority I know of who accepts ghostwriting defined in this way. The ICMJE calls for acknowledgments. So, as rethoryke wrote the “lead authors are not doing the bulk of the actual text production, but _did_ run the study” is “different from a case where the lead authors were not involved with the data collection or the writing, but instead are figureheads for a manuscript prepared before they were ever approached.”

And to the final “G” word, guest authoring. (I’m counting these two words as one.) This occurs when authors are named in the byline when they do not meet ICMJE criteria for authorship. In the real-world example, In that commonly discussed example I referred to before, the medical education and communication company offers authorship to “key opinion leaders” (prominent clinicians and researchers) to sign on as authors.

In the example commonly thought of, ghostauthoring (on the part of the pharmaceutical company and the medical education and communications company) and guest authoring (on the part of the key opinion leaders) happen together. A medical writer is involved to write up the material. If the writer is not acknowledged, then that is ghostwriting; but if the writer is acknowledged, then that is not ghostwriting.

So now, Rodney Wilson, you have an idea there are ghostauthored studies. The checks and balances repose in the consciences of staffs of pharmaceutical companies and of medical education and communication companies, key opinion leaders, and freelance medical writers and editors (like me.)