When Dr. Martha Gerrity accepted a position last summer as co-editor of a small but respected medical journal, she understood that part of her job would be blocking attempts by drug companies to influence the scientific literature.
She just didn’t know how big a part of her job it would be.
Gerrity, a physician at the Portland VA Medical Center, is co-editor of the Journal of General Internal Medicine. JGIM is a small, peer-reviewed journal that publishes studies and reviews that help health-care professionals determine how to treat their patients, and with which medications. It is unusual among medical journals in that it does not accept advertising from the pharmaceutical industry. Almost all of JGIM’s subscribers are academic physicians who are responsible for educating medical students.
Gerrity said she has come across four instances of what she termed hidden conflicts of interest this year. One of the conflicts, she said, raised the issue of whether a drug company hired ghostwriters to shop around a review article favorable to the company’s commercial interests. The drug company involved denies the journal’s version of events and suspicions.
That particular conflict is the subject of an article and editorial published in JGIM’s online version.
The case study sparked a discussion that has helped to persuade the World Association of Medical Editors to tighten its policies on ghostwriting.
The article focused on the widely used oral blood-thinner warfarin and how it interacts with herbs and dietary supplements. The subject seemed worthy of review, Gerrity said, so she and her colleagues forwarded it to peer reviewers with relevant expertise.
One of the reviewers, Dr. Adriane Fugh-Berman of Georgetown University, immediately contacted Gerrity and her co-editor, Dr. William Tierney of Indiana University, with the allegation that a company working for the British drug giant AstraZeneca had tried to sign her on as author of the same article. Fugh-Berman said she turned down the offer because it seemed biased against warfarin with the obvious commercial goal of creating a market for a new drug being prepared by AstraZeneca.
The conflict of interest statement on the article submitted to JGIM mentioned “editorial assistance” from the British firm Rx Communications Ltd. and an “educational grant” from AstraZeneca.
Gerrity said that statement was incomplete. “We had no idea who had done the research,” she said. “It wasn’t clear what the funding relationship was between the pharmaceutical company and the medical education company.
“But the more important issue is the intent,” she said. “The intent was to bias the medical literature in favor of a pharmaceutical company’s product. … This isn’t telling the truthful story about warfarin.”
The new AstraZeneca drug, ximelagatran, has thus far been denied approval from the Food and Drug Administration because researchers have found it could harm a patient’s liver.
Writer’s identity protected
Fugh-Berman informed Gerrity and her co-editor that in her view the paper was “transparently biased” because it omitted some controlled trials and highlighted other data she considered less valuable.
Tierney wrote the author to explain that the article had been rejected and that he considered the submission an egregious breach of ethics.
JGIM’s privacy rules require that the journal must protect the identity of the doctor who submitted the rejected article. The author did not receive any drug company money, according to Rx Communications, although published articles are crucial to the careers of academicians. Nor did the journal editors name AstraZeneca or RX Communications in its article and editorial or in interviews with the Portland Tribune.
Gerrity said they treated the situation with caution for practical reasons. “As a small, specialist society, we don’t have the resources to defend ourselves, even though we would probably come out fine in a court of law,” she said.
Besides, she added, “it’s not just this one company. This is a bigger issue.”
After receiving Tierney’s letter, the author of the submitted article and the president of Rx Communications wrote back to say that there actually were two papers on warfarin in development, and the paper submitted to JGIM was indeed the work of the named author. According to their explanation, portions of the article Fugh-Berman was later asked to review were mistakenly sent to her when she was solicited to write a different article.
Ruth Whittington, president of Rx Communications, wrote to Fugh-Berman to “apologize sincerely for the distress caused to you and the journal editor by our mistake.”
Drug firm explains
AstraZeneca’s public relations director, Julia Walker, said the article sent to JGIM conformed with the company’s publication policy that requires named authors to “be fully involved from the onset … offer substantial contributions to the conception, acquisition or interpretation of the data” and “retain responsibility for the article’s content.”
Walker said that while the article’s named author received assistance from a medical writer, she did not merely put her name on top of an article that already had been prepared for AstraZeneca. “The named author provided significant contributions over many months to the content and focus of the (article),” Walker said.
In a written statement e-mailed to the Portland Tribune on Thursday, Whittington said: “We can confirm that none of the articles mentioned by Dr. Fugh-Berman were ghost-written. In error, we sent out an advanced draft written by one of our authors to Dr. Fugh-Berman. We apologised for this error some months ago, and an explanation for how it occurred was given to Dr. Fugh-Berman, the JGIM, the author and to AstraZeneca.”
Fugh-Berman called the explanation “ridiculous.” She also questioned whether AstraZeneca was sticking to its editorial policy to require that the named author be involved from the beginning, since the article she was asked to write came to her not as an outline but as a completed 2,850-word manuscript, including 65 references and a title page with her name already on it.
Fugh-Berman said there were few differences between the article she received and the article she was later asked to review. She said that in her opinion some of the changes indicated a bias against warfarin.
Tierney said the warfarin article led JGIM to sharpen its editorial policies regarding ghostwriting. Anyone submitting articles for publication must list all contributors by name, including researchers who compiled data by reviewing articles. Authors also must spell out all financial relationships with commercial interests.
Collaborations can be OK’d
“We don’t rule collaborations out,” Tierney said. “We just ask people to make it explicit what their relationships are with the drug companies. We will make sure that that information is available to the reviewers. And if it still looks like a good article, we’ll publish it.”
Tierney argues that collaborations between drug companies and academic physicians are inevitable and even necessary, because medications need to be carefully tested by trained researchers to ensure that they are safe. He himself has co-authored several papers with drug company researchers.
But allowing drug companies to masquerade marketing messages as science is another matter, Tierney and Gerrity argue in their JGIM editorial: If biased studies slip through the peer review process and make it into print, “it injects bias and untruth into the scientific dialogue in order to enhance corporate profits.” They argue that the only way to prevent such abuse is to require complete transparency from all named and unnamed authors involved in preparing a manuscript.
The World Association of Medical Editors agrees and has decided to tighten its policies regarding authorship and hidden funding following an involved online discussion initiated by Tierney.
Many editors from medical journals wrote of similar ethical quandaries in the discussion. Still, the scope and seriousness of the ghostwriting issue remain a matter of debate.
Fugh-Berman, who speaks regularly at national medical conferences, said many of her colleagues are approached to claim authorship of studies already ghostwritten by writers hired by companies working for pharmaceutical companies. Some researchers are paid handsomely, she said, while others use the publications to advance their careers in the competitive “publish or perish” academic environment.
Fugh-Berman is convinced that many articles by what she calls “sham authors” make it into print. This is a problem because “once something is published in a medical journal, it is citable as fact forever,” she said.
Others in the field question whether the practice is common.
“I have absolutely never heard of that in my life,” said Patricia Garrison, a medical writer based in New York.
Garrison said that in her experience, the doctor is always contacted before the article is written, and the medical writer merely acts as a wordsmith or editorial assistant to a physician who doesn’t have time to spare.
Parliament looks into studies
A new report prepared for the British House of Commons investigating the influence of the pharmaceutical industry found that there was nothing wrong with a writer helping a busy doctor to complete a study. But it raised concerns that some ghostwritten articles paid for by drug companies are misleading and biased because they might select data that is favorable to their commercial interests while leaving out details that might jeopardize those interests. Such articles are particularly problematic when researchers sign on as primary authors without seeing the raw data that lies at the article’s foundation, the report found.
During the British parliamentary hearings in January, Dr. John Patterson, executive director of development for AstraZeneca, testified before the Health Committee of the House of Commons that ghostwriting “is absolutely not practiced in my company.”
Gerrity said that since receiving the warfarin article she has discovered three additional hidden conflicts in papers submitted to JGIM. One study was rejected as poorly done; another was well-done but biased toward a drug company that provided funding; the third paper was sent out to peer reviewers who “saw right through it,” she said.
Gerrity said her experience with the warfarin article has made her more vigilant about requesting precise explanations regarding potential conflicts of interest, including honorariums that some doctors consider their own private financial business. That hasn’t made her job any easier, but Gerrity says she’s determined to stick with it.
“We’re hoping to get people to rethink the whole practice of seeding the medical literature with biased studies,” she said.
http://pharmawatch.blogspot.com/2005/04/article-by-ben-jacklet-in-portland.html
