© Health Care Services, California
"It is as if someone sat at a desk and wrote a novel about a research idea," the committee wrote in a 29 June summary report posted in Japanese on the society's Web site.
The fabrications could produce a record number of retractions by a single author if the journals, as seems likely, decide to retract the papers. ScienceInsider was unable to reach Fujii, who had asked the society not to provide the media with his contact information.
Fujii's findings have been under a cloud since 8 March when an analysis in the journal Anaesthesia raised questions about his data. On 9 April, 23 journal editors publicly asked seven Japanese institutions named in the papers to investigate. The anesthesiology society took on the task because "it would have been difficult for any one institution to clarify what happened," says Koji Sumikawa, an anesthesiologist at Nagasaki University who headed the investigation.
The panel focused on 212 of 249 known Fujii papers. It tried to review the raw data, laboratory notebooks, and records on the patients or animal subjects involved. Committee members also interviewed relevant people.
Among the 172 papers judged bogus, the report claims that 126 studies of randomized, double-blind, controlled trials "were totally fabricated." The committee identified only three valid papers. For another 37 papers, the panel could not conclusively determine if there had been fabrication.
The report states that Fujii started falsifying data in 1993 and found a pattern of fabrication that stretched through his successive stints at Tokyo Medical and Dental University, the University of Tsukuba, and Toho University in Tokyo. Toho dismissed Fujii last February when an internal investigation found he never got ethical review board approval for several studies.
Sumikawa says the panel concluded that Fujii tried to hide what he was doing, even from his co-authors. He deliberately blurred the timing of the studies and where they were conducted, the report states, by naming hospitals and institutions where he worked part-time or had some affiliation.
The panel said that the responsibility of those co-authors ranges from "serious" to "none at all." The only one of Fujii's co-authors specifically named in the summary is University of Tsukuba anesthesiologist Hidenori Toyooka.
The report says Toyooka "was not involved in fabrication but bears significant responsibility" since he was Fujii's supervising professor both at Tsukuba and when they both worked at Tokyo Medical and Dental University. Toyooka is listed as a co-author of many of the papers cited by the 23 journal editors. (Toyooka could not be reached for comment.)
At the same time, the investigation found that some scientists were unaware Fujii had included them as co-authors. In one case, two supposed co-authors told the panel their signatures on a submission cover letter were forged. A post from 18 June on the blog Retraction Watch suggests why the researchers didn't realize their names were on Fujii's papers: The papers didn't attract much attention. The blog notes that three recently retracted papers only garnered six, four, and three citations.
Despite the low impact of the work, Fujii apparently used his high productivity to land new jobs, obtain public research funding, and garner fees for speaking at industry seminars, according to the panel's summary report. He even applied for prizes offered by the society, although he was never chosen.
Sumikawa says the summary is being sent to the 23 editors as well as to the institutions involved. Formally, it is the responsibility of those institutions to request retractions. Responding to a query from ScienceInsider, the University of Tsukuba's public relations department issued a written statement indicating the school is taking the society's findings seriously and will consider them in its own ongoing investigation.
German anesthesiologist Joachim Boldt is believed to hold the dubious distinction of having the most retractions - about 90. Boldt's scientific record also came under fire several years ago by some of the same journal editors questioning Fujii's work.
Sumikawa says Fujii recently contacted the society claiming that parts of his work were valid, "but he didn't identify which parts." Sumikawa says the society plans shortly to post a detailed report in Japanese, followed by an English translation.
The committee recommended several measures aimed at reducing the incidence of misconduct. They include regular seminars to raise awareness of the ethical aspects of medical research and to clarify the responsibilities of lead authors and co-authors. The report also said scientific societies, along with journals and institutions, should be prepared to investigate when questions are raised about publications by their members.
*This item originally stated that a 29 June post on the blog Retraction Watch suggests that Fujii's fake co-authors didn't realize their names were on papers because the papers didn't attract much attention. That post was from 18 June and the 29 June post discussed the number of papers in which Fujii is said to have faked data.
Reader Comments
to our Newsletter