The University of Pennsylvania has found no merit to an allegation that two of its psychiatry faculty committed research misconduct when they co-authored a 2001 paper written with help from writers hired by a drug company. But the university says that under current rules, the two faculty members would have been required to acknowledge the role of a "medical writer" in preparing the paper.
Last July, Penn psychiatrist Jay Amsterdam alleged in a letter to the federal Office of Research Integrity that five researchers, including Penn's Laszlo Gyulai and Dwight Evans, chair of the Penn psychiatry department, had "engaged in scientific misconduct by allowing their names to be appended to a manuscript that was drafted by" Scientific Therapeutics Information (STI), a medical communications company, that had been "hired by" GlaxoSmithKline (GSK). The paper, which appeared in June 2001 in the American Journal of Psychiatry, reported on a small clinical trial of the antidepressant Paxil, partly funded by GSK and the National Institutes of Health. Amsterdam also claimed the paper was "biased" in favor of Paxil's safety and efficacy.
Amsterdam's letter argued that ORI should be involved because NIH Director Francis Collins has written that ghostwriting "may be appropriate for consideration as a case of plagiarism," which falls under the federal definition of research misconduct.
But Penn has concluded that no plagiarism occurred. In a statement yesterday, the university says that a faculty committee found "there was no plagiarism and no merit to the allegations of research misconduct" because Evans and Gyulai helped conduct the research and analyze the results and "contributed to the paper." The paper "presented the research findings accurately," Penn says.
The statement goes on to say that "with respect to the allegations of ghostwriting," the committee looked at whether "the medical writers engaged by the study sponsor" should have been acknowledged in the paper. While medical journals and Penn now require this, "guidelines in place in 2001 did not." The statement also says that the affiliations of the authors (including three GSK employees) was included in the original manuscript, but the journal removed the information. And the Penn review committee dismissed Amsterdam's claim that he should have been a co-author, not just listed in the acknowledgements, because his role "did not meet the journal's guidelines for authorship."
According to the Pharmalot blog and The Chronicle of Higher Education, Amsterdam's lawyer plans to bring the case to ORI's attention again and to Senator Charles Grassley (R-IA), who has investigated conflicts of interest in medical research in the past.