By Andrew M. Seaman
NEW YORK (Reuters Health) - Most major medical journals don't spell out how they handle the omission from a published study's author list of a person who substantially contributed to the work, according to a new study from Spain.
So called ghostwriting raises concerns about the validity of research findings because it can involve researchers submitting as their own work an article that was largely drafted by a pharmaceutical company's paid writers, for example.
Of 399 peer-reviewed medical journals the new study's authors analyzed, they found that only one in ten defined ghostwriting and even fewer had written policies on how they detect or respond to the practice - considered by many to be misconduct.
"The kind of ghostwriting I'm concerned about is where a corporate author - usually a medical writer employed by a medical company with a conflict of interest - is not listed on the byline," said Jeffrey Lacasse, who has researched ghostwriting but was not involved with the new study.
Lacasse, from Arizona State University in Phoenix, said he is not surprised by the study's findings, but they are concerning.
"We're talking about scientific integrity," he said.
In 2011, researchers from the Journal of the American Medical Association published results from an anonymous survey that asked 622 authors of medical research if there was a ghostwriter on work they published in 2008.
Almost 8 percent of the authors surveyed said there were ghostwriters on their articles. That was down from almost 12 percent in a similar 1996 survey.
The International Committee of Medical Journal Editors (ICMJE), a group that recommends best practices for journals, says study authors should "have participated sufficiently in the work to take public responsibility for appropriate portions of the content."
The new study's lead author, Dr. Xavier Bosch from the University of Barcelona, and his colleagues write in JAMA Internal Medicine that journals have led the way in creating editorial policies for scientific publication, but no study has looked at whether those policies address ghostwriting.
For their analysis, they examined the websites of 399 peer-reviewed, English-language medical journals to see if their editorial policies defined or addressed the editors' procedures for detecting or responding to ghostwriting.
Overall, ghostwriting was mentioned in about 17 percent of the journals' policies, but only one in ten actually defined it. Only 4 percent spelled out how they detect ghostwriting and about 6 percent explained how they respond to it.
But Dr. Christine Laine, editor-in-chief of the Annals of Internal Medicine and member of the ICMJE secretariat, told Reuters Health that she worries the new study left out some journals that address the practice without using the terms "ghostwriting" or "ghostwriter."
"I have concerns that they are misclassifying journals that do have policies against ghostwriting, but not in the narrowly defined method of that study," she said.
For example, she said ICMJE does not mention the term "ghostwriting" in its requirements, but it's implicitly addressed through the criteria the organization sets for authors.
"My concern with the study in JAMA (Internal Medicine) is that they looked at those terms rather than the concept," Laine said.
Lacasse, however, said ICMJE's criteria still allows ghostwriters since it says authors need to give "final approval of the version to be published." He said the other authors would just not ask the ghostwriter for final approval.
"I think there are some journals doing a better or worse job. But as you can see from the letter, we still have a tremendous problem on our hands," Lacasse said.
The study researchers could not be reached for comment.
JAMA Internal Medicine's consulting editor Catherine Nancarrow wrote in a note accompanying the new study that authors are required to be transparent and agree to standards similar to ICMJE's recommendations when they submit articles to her journal.
"They serve as an important first step to ensuring an honest and open relationship among authors who entrust their work to a journal; the journal whose privilege it is to deliver it to the global community; and our readers, who can be assured of the integrity of the information and findings reported," she wrote.
Laine said she believes a cultural shift within the medical community has improved the conduct of researchers.
"I'd like to think that people are learning from the mistakes of others. And I think it's just a whole other level of scrutiny and transparency around conflicts of interested in biomedical science that maybe is making everybody behave a little better," she said. "At least the optimist in me wants to think that."
SOURCE: http://bit.ly/10EUiDF and http://bit.ly/16Mwbng JAMA Internal Medicine, online April 8, 2013.