How I was asked to 'author' a ghostwritten research paper. 
By Adriane Fugh-Berman 

Thursday April 21, 2005
 Source The Guardian 

Recently, the House of Commons health select committee looked at the submission of ghostwritten articles to medical journals. Witnesses from two pharmaceutical groups, GlaxoSmithKline and AstraZeneca, "strongly denied that ghostwriting was practised in their respective companies". 

But such work can be hired out. Last summer, I was asked by RxComms, a British medical communication company, to author a review of interactions between herbs and warfarin (a generic anticoagulant prescribed to prevent strokes or blood clots). Well, not "author", exactly. The usual practice is for a complete article to be supplied; all I would have to do was review it and sign it off. 

Months later, I received a completed, 2,848-word draft, with an abstract, references, and a table, ready for submission to a journal, with my name on it. A note asked me to return it with any changes within seven days. 

I asked why AstraZeneca, sponsor of the article, was funding a manuscript that mentioned none of its products, and I was informed by RxComms that the paper was part of a series meant to highlight problems with warfarin - in particular, "warfarin's high interaction potential, which can give rise to problems with anticoagulation control". It seemed to me that the article was intended to help AstraZeneca lay the groundwork for a new drug, ximelagatran, to compete with warfarin. 

Many articles in medical journals are ghostwritten to order for drug companies, often by writers for medical communication companies, who appear to be acting as intermediaries to distance drug companies from the articles. Faculty members are recruited to have their names appear as authors. The arrangements are made with the communication companies. 

With ghostwritten articles, the author may opt to contribute, but if changes are not advantageous to the sponsoring company, the article may be ditched. We don't know how many of these articles infest medical journals, unbeknown to their editors. Most bona fide journals require authors to disclose conflicts of interest, but editors cannot enforce honesty. 

Drug advertising to physicians is aimed at influencing prescribing, but at least is a recognisable form of persuasion. More insidiously, drug companies sponsor many talks at medical meetings or conferences. In the US, it is so common for a lecturer to be in the stable of a drug company that, when invited to speak, I am often asked which firm usually sponsors my talks. 

I declined the offer from RxComms, but another "author" was willing to do it. A few weeks later, a manuscript with alarming similarities was sent to me for peer review by the Journal of General Internal Medicine. On being told I believed that the paper was ghostwritten, the journal editors rejected it, told the "author" not to submit a paper again, and informed the World Association of Medical Editors. RxComms says this was a different manuscript and was actually written by the person who submitted it, but it had been sent to me in error. AstraZeneca has also denied that the article was ghostwritten. The company says it has strict guidelines, insisting that authors make substantial contributions to these kinds of articles and take responsibility for the contents: "Most pharmaceutical companies, including AstraZeneca, use professional writers to assist in manuscript development, when the named authors lack the time or expertise to produce a well-written publication." 

I have found this practice is standard within the industry. "Publication development" is an important part of marketing, and entire conferences on the subject are held. Conference talks may promote available drugs, or be part of a campaign for medications not yet available. Drug companies spend three times as much on promotion before a drug is launched than in the year after launch. These efforts may include sending "key opinion leaders" on speaking tours and seeding journals with articles that stress the importance of a targeted disease, or problems with a competitor's drug. 

Medical education companies are often hired by drug companies to put together conferences, some of which I have spoken at. Filler talks such as mine on herbs and dietary supplements act as cover for message talks that advance the sponsor's goals. "Message talks" are not obviously biased and often reference the targeted drug merely as one of several choices. The more unbiased a talk seems, the more useful it may be. 

Doctors trust articles in journals and rely on the experts whose names adorn them. If they do not know the origin of those articles, then the system needs reforming to eliminate both ghost and guest authors, and to ensure that marketing pieces cannot masquerade as academic articles. 

· Adriane Fugh-Berman is associate professor in the department of physiology and biophysics at Georgetown University School of Medicine