Ghostwriting dispute spooks researchers

Controversial practice could compromise integrity of research

Recent allegations that Merck used ghostwriters to write articles about Vioxx have left some experts saying that the practice could compromise the integrity of the scientific research system.

These experts contend that ghostwriting – when a company fails to credit an author who had a substantial role in shaping an article or the research behind it – chips away at a system intended to ensure that safe and effective drugs and devices are brought to market.

When researchers are listed as authors of a scientific article, the expectation is that each has had a hand in independently framing the research question, designing the study, objectively evaluating the data and reporting the data accurately.

But problems arise when a pharmaceutical company develops a clinical trial in an effort to win approval for a drug, funds and oversees that trial and then forwards a manuscript to an author to either sign off on it or make only slight revisions, said Jerome Kassirer, a physician and former editor of the New England Journal of Medicine.

However, experts noted that ghostwriting shouldn’t be confused with a situation where a talented writer assists a talented scientist in turning his research into a journal article – as long as the writer’s contribution is acknowledged.

“I see no problem in that,” said Marc Rodwin, a professor of law at Suffolk University Law School who has written extensively on conflict-of-interest issues.

But problems do arise, he said, when an article is written “independently of the person who puts their name on it, particularly when [the research is] driven and funded by a pharmaceutical firm and then handed to someone to sign it.”

The latest incident came to the forefront in April when two articles in the Journal of the American Medical Association alleged that Merck hired ghostwriters to pen articles about Vioxx studies.

The Journal claimed that those articles were later signed by researchers who had little to do with the studies and reported incorrect data minimizing the risk of death associated with taking Vioxx.

Merck took Vioxx off the market in 2004, and since then the company has battled litigation over the deaths of patients who took it.

In a statement, Merck responded that JAMA’s articles contained comments that are “false, misleading or lack context” and said it wasn’t “given an opportunity to respond prior to publication.”

Further, Merck contended that the articles were based on analyses performed by consultants who were hired by trial lawyers and made “allegations similar to those previously advanced … at trial.”

The blame game

The latest ghostwriting allegations are part of a string of cases dating back to 1959. The U.S. Senate held hearings on similar matters in 1962, Rodwin said.

Experts blame a wide range of parties for the flaws of ghostwriting.

A scientist who agrees to put his name on an article even if he hasn’t “seen the raw data or senses the material has been prepackaged or prewritten by a company” violates a longstanding academic tradition and expectation that authorship is “meaningful,” said Thoru Pederson, a biomedical scientist and the associate vice provost for research at University of Massachusetts Medical School.

He asserted that an author should also be aware of the raw data from concurrent studies taking place at other sites that will be published under a joint byline.

Rodwin said that ghostwriting is deceptive because it leaves both the researcher and the company pretending they completed independent analysis of the drug’s safety. He contended that ghostwriting can rise to the level of fraud in some instances.

Pederson said that medical journals that publish the studies without a rigorous examination of bias or a proper amount of skepticism are also to blame.

“I don’t think JAMA and the New England Journal of Medicine have been aggressive enough” in fighting this problem, he said.

A key red flag is that many of the corrections published by journals deal with authors’ names left off articles. That’s a sign that the articles were never circulated among the researchers who worked on the studies, and therefore that those authors never saw the final paper.

In the recent case, JAMA editors took responsibility for the role they played in disseminating the original articles about Vioxx in an editorial in the April 16 issue. That editorial also called for reform within the publication system.

Simple solutions

Simple changes at all points in the process from investigating the safety of a drug to publishing an article about it can help stave off these situations.

Kassirer advised that researchers insist that they write a manuscript on their data themselves and have final control over everything the manuscript says.

They should also ensure that the drug maker has no input on the manuscript whatsoever, he said. Such arrangements must be made with the company before the study begins.

The same holds for review articles, which a researcher should generally write him or herself.

And if a situation arises where “a company does provide a manuscript or a review article for [a researcher], then [he or she has] to go over every line, every word … and change the manuscript to reflect [his or her] own point of view, not the company’s point of view,” he said.

These requirements are particularly important “when it’s clear it has to do with … pushing something to get on the market,” Rodwin said.

Journals often require and publish statements about which authors conducted what work, Pederson explained, making it easier to identify authors who claimed to have a role they did not actually have.

Journals should also “insist that they know whether the information has been in the hands of the investigator all the time and whether the writer takes full responsibility for the manuscript,” Kassirer said.

Read with skepticism

Doctors and researchers must carefully consider the conditions surrounding the clinical trials they read about in medical journals and approach the results with a healthy degree of skepticism.

Readers must keep in mind the commercial context in which these articles are written and in which drugs are investigated.

“I think most physicians and scientists, when they read articles, don’t think of it in the commercial context to the degree they should,” said Pederson.

But Rodwin cautioned not to take that approach too far because it’s impractical for a reader to study an article with a checklist of red flags beside it.

“Any … professional has to weigh information and make the best clinical decisions,” he said. “Good doctors are doing that all the time. The point is to … not assume you need to conduct an independent review of how something came to market.”

Questions or comments should be directed to the
editor at: reni.gertner@mamedicallaw.com