Whether or not you are familiar with his work in condensed matter physics, you have probably heard of Jan Hendrik Schön. Unfortunately, Schön is famous not because he advanced his discipline, but rather because he was caught committing that most noxious of scientific sins: falsifying data.
Although an independent review committee determined that Schön did indeed perpetrate fraud, his co-authors were cleared of this charge. According to the report released 25 September 2002 by the committee, "There is no implication here of scientific misconduct [on the part of the co-authors]; the issue here is one of professional responsibility." The committee goes on to state that co-authors are not responsible "for the entirety of a collaborative endeavor."
Data falsification and fabrication occur more frequently than the scientific community would like to admit. And although everyone readily concedes that such practices constitute egregious scientific misconduct on the part of the perpetrator, to what extent does responsibility for tainted papers reside with the co-authors?
The International Committee of Medical Journal Editors has proposed that each person's contributions be listed in the published paper. A few journals--such as the Journal of the American Medical Association (JAMA)--have implemented this requirement, but JAMA represents the exception and not the rule.
That is partly because there is no consensus on the issue among the professional societies that put out many of the scholarly journals in which you may eventually see your research published. Authorship ethics policies range from the American Chemical Society's broad stance that "the co-authors of a paper ... share responsibility and accountability for the results" to the American Physical Society's (APS's) Ethical and Values Statements, which say that "all collaborators share some degree of responsibility for any paper they co-author." The APS statement goes on to explain that some authors may be held responsible for an entire paper, but those scientists who make narrower contributions "may have only limited responsibility." The American Society for Cell Biology, according to Stephanie Dean, its director of publications, currently does not have a policy on author responsibility.
This is not an issue that is going to go away. Collaborations are becoming increasingly common in science, particularly those involving interdisciplinary teams. It is difficult enough to monitor data being churned out of a collaborator's lab when you both work in the same field, but what if you are a biologist collaborating with a physicist?
Everyone wants to be given credit for the work they do, and authorship is widely viewed as the ultimate scientific credit. Indeed, in Next Wave's previous examination of authorship ethics, we focused on the criteria that ought to determine who is included as an author and who is not. Now, however, we turn our attention to scientists' responsibilities once they are on the author list and the paper is published.
As before, we have asked a panel of experts and interested parties to peruse the following fictitious case study and address the complex issue of co-author responsibilities in a responding essay.
William Lowe is a postdoctoral associate in an organic chemistry lab at a well-known institution. The laboratory consists of two postdoctoral associates and six graduate students. William and his PI, Sara Bridges, have been trying to develop some research ideas focusing on the development of pharmaceuticals to treat HIV. They finally come up with a project idea for one of the first-year graduate students, Terry Jones, to pursue, and they decided that Terry would conduct his research under William's guidance. After 1 year, Terry successfully synthesized a potential HIV drug that showed promise when used in a rat model. All the lab members were excited about the dramatic results. Terry prepared a manuscript for publication, and a second postdoc in the group, Tina, edited the paper prior to submitting it to Sara for review. It was decided that because Terry conducted the experiments, he would be first author. William and Tina were listed as co-authors, as were Mary Greene and Toby Marshall--two PIs from other institutions who contributed research materials. Sara was listed as the senior, corresponding author. The paper was submitted to a well-respected peer-reviewed journal, which published it 4 months later.
It seemed that everyone was excited about this potential major breakthrough for HIV research. Sara even became a household name after she was asked to appear on major news shows to discuss her lab's findings!
During this media whirlwind, representatives of a major pharmaceutical company contacted Sara about the possibility of collaborating. Sara felt that this was a great opportunity to develop a partnership with industry. The company wanted to conduct clinical trials on humans with the chemical, but first, it wanted to replicate the original work. Unfortunately, when the company synthesized the drug and used it on rats--just as Terry did--there was no effect! Bob Smith, the head scientist at the company, requested Terry's notebooks and found that many data entries had been written in pencil, then erased and written over. Bob could decipher some of the earlier entries, which actually indicated that the chemical was not having a positive effect. He began to suspect that Terry made up the positive data. ...
Drummond Rennie, deputy editor of JAMA, says that some of the players in the scenario shouldn't have been authors in the first place. Based on how similar real-life cases have played out, he predicts that the co-authors will plead ignorance and seek to "distance themselves from Terry."
Instructor Christina Ochsenbauer agrees that the contributions of some of the co-authors did not warrant their inclusion on the paper. Ochsenbauer also suggests that the "lessons from the case study apply to any young investigator on the brink of [their] independence."
W. F. Brinkman, president of APS, wonders whether the co-authors should have become suspicious of Terry's data much sooner. And he acknowledges that authorship remains under active debate within APS.
Graduate student Sue Buckingham explains her view on how the misconduct investigation should proceed. She also says that although Sara may have "placed too much confidence in William to oversee Terry's research," Terry is still an adult. PIs shouldn't have to "micromanage" their students, she thinks.
Matt Maneen, a graduate student at the University of Vermont College of Medicine, believes that "the ultimate responsibility resides with the PI." Terry may have fabricated data, but it's Sara's laboratory and reputation that are on the line.