Harvard Medical School is investigating plagiarism allegations brought against one of its clinical professors of rheumatology after a search tool detected a possible instance of duplicated work.
Professor Lee Simon's 2004 review of rheumatoid arthritis was retracted on Jan. 29 from the journal Best Practices & Research Clinical Rheumatology. The journal's publisher was notified that Simon's work might have been plagiarized after a scanning program developed by researchers at the University of Texas Southwestern Medical Center discovered similarities between Simon's research and an earlier paper by Roy Fleischmann, a professor at UT Southwestern.
Researchers detected that Simon seemed to have copied entire pages of text without citation from the journal Expert Opinion of Drug Safety, which published Fleischmann's article in 2003.
Harvard is in the early stages of looking into Simon's case, examining whether plagiarism or scientific misconduct was involved. The medical school has formed an internal ad hoc committee to review the plagiarism accusations.
David Cameron, associate director of public affairs at Harvard Medical School, stressed that the committee has not come to any conclusions.
"This is extremely preliminary," Cameron told the Daily. "We are in the very early stages of looking at this accusation."
Cameron outlined the types of hypothetical questions the ad hoc committee could pose in its investigation. "What is the information in this paper that appears in other sources?" he said. "Is this the kind of thing that is commonly done in this type of paper?"
The committee may determine that the allegations of plagiarism are unsubstantiated, in which case the inquiry will cease. But if the committee decides that there is a basis for scientific misconduct, a formal investigation will be launched with external participants and stricter guidelines.
Elsevier, the publishing company of the journal that published and then retracted Simon's paper, declined to elaborate on the retraction. But the company did say that it was leading its own investigation into Simon's work.
"Elsevier takes concerns about plagiarism and ethical misconduct in publishing very seriously," Tom Reller, the company's director of corporate relations, told the Daily. "We are currently investigating this matter in accordance with our usual procedures."
Mounir Errami, a research scientist at the UT Southwestern Medical Center, explained how eTBLAST, the copycat detection program developed by the team from UT Southwestern, discovered Simon's alleged plagiarism.
"[The program] compares the citations from the major biomedical database that's called Medline," Errami told the Daily. "It found a number of cases where the citations were very close to each other."
Errami said eTBLAST has been available since 2003, when it was originally designed to search for academic literature by topics.
The program was modified last year to identify instances of duplication in biomedical papers by comparing their citations, titles and abstracts within Medline. There are about 17 million papers included in the Medline database, but UT Southwestern researchers have only scanned seven million of them to detect duplication.
The research team ran these papers through the eTBLAST algorithm last fall and came up with about 80 cases of duplication by different authors that had a high probability of plagiarism. The papers had to be manually verified, and in 10 cases the team found it necessary to notify the authors and publishers of the original papers and duplicate papers of their findings. eTBLAST heard back from eight of these 10 publishers. Simon's article was one of these cases.
"It was late November, and we did about one case a week," Errami said. "Simon was one of the first ... We know that out of these eight cases, three have official investigations, one of which is the Simon case."
He explained that members of the UT Southwestern team do not refer to their duplication findings as plagiarism because they are "not allowed to make a judgment call."
"But these [cases] are very questionable," he said.
He explained that the team plays an identification role only, bringing facts to light and then stepping out of the picture.
"We ask questions, and from there we are totally excluded from the process," Errami said.
Medline citations that show two remarkably similar papers are compiled in the database Déj? vu, which is available to the public online. This database can be used to detect possible cases of plagiarism.
