We need to identify consequences sufficient to deter the unscrupulous few whose deeds shadow the accomplishments of all diligent and honest scientists.
Two seminal articles published in 2004 and 2005 in the prestigious scientific journal Science heralded a new era in stem cell research. Findings from the study published in November 2005 pointed to cures for millions of people with diseases ranging from diabetes to paralysis resulting from spinal cord injuries. This study generated significant optimism among patients and was reported widely in both the scientific and public media.
Then on Jan. 10, 2006, Seoul National University announced that South Korean scientific icon veterinarian Dr. Hwang Woo Suk was responsible for fabrication and manipulation of the data presented in the two articles. Science immediately retracted both articles, but not until more than 200 other published scientific articles already had cited Dr. Hwang’s findings.
Just three days after the stunning announcement from South Korea, another prestigious journal, The Lancet, learned that data it published in an October 2005 article may have been "manipulated." The next day, representatives from the Norwegian Radium Hospital Oslo announced that not only had Dr. Jon Sudbø—a researcher on their staff and the lead author of the Lancet article—manipulated the data, but his study was a "complete fabrication."
The study purportedly looked at the incidence of oral cancer and cardiovascular disease among people using nonsteroidal anti-inflammatory drugs (NSAIDs). Results of the study suggested that the use of NSAIDs reduced the incidence of oral cancer, but increased the risk of death from cardiovascular disease.1 Of the 908 individuals said to be part of the study, the author apparently fabricated data for 454 people with oral cancer. The Lancet retracted the article.2
Then in February 2006, The New England Journal of Medicine printed an "expression of concern" regarding two additional publications authored by Dr. Sudbø. In one of these articles, Dr. Sudbø reported that 26 of 27 patients with aneuploid mouth lesions developed aggressive oral cancer and were highly likely to die from their disease. Another study by Dr. Sudbø, this one published in the Journal of Clinical Oncology, also may contain fraudulent data and is being reviewed together with all of his other published studies.
It is never easy to detect and expose scientific fraud. Sometimes we learn of it through whistle-blowers, usually junior members of a research team, who risk retribution and alienation by the scientific community. The best defense against fraud is the scientific principle of reproducible results. Misgivings about a study sometimes are raised when results of that study cannot be replicated. However, it may not always be financially feasible to replicate a study costing many millions of dollars.
The responsibility for the accuracy of research presented in a manuscript rests solely with the author. Peer reviewers can assess the analysis of the data as submitted and evaluate the significance of the study as presented in the manuscript. Unless it is obvious, however, the peer review process seldom discovers scientific misconduct. Original data usually are not available to the reviewers, and journals do not have the time or resources to investigate every submission.
When a journal publishes a manuscript, it lends validity to the content and, therefore, to the actual research presented in the article. At that point, some of the responsibility for the findings shifts from the author to the editor. When scientific misconduct, such as fabrication or manipulation of data, comes to light, the journal has a responsibility to make this known to the research community. Otherwise, the journal in question perpetuates the fraud by letting future research build on fallacious science.
It is not always clear how to handle allegations of fraud and misconduct. It is the responsibility of academic or government-sponsored agencies to conduct the necessary investigation, not scientific journals.3 However, journal guidelines of ethical conduct could include sanctions, such as preventing future publication of material from the offending authors. Disciplining authors, notifying outside funding sources or even pressing criminal charges is not within the scope or responsibility of a scientific journal. For an example of what the penalties for scientific fraud might be, consider the case of Eric T. Poehlman, PhD. The following are excerpts from a media release issued by the U.S. Department of Health and Human Services Office of Research Integrity on March 17, 2005.4
"Dr. Eric T. Poehlman, 49, a former tenured research professor at the University of Vermont (UVM) College of Medicine in Burlington, Vermont, has agreed to a comprehensive criminal, civil, and administrative settlement related to his scientific misconduct in falsifying and fabricating research data in numerous federal grant applications and in academic articles from 1992 to 2002.
"... Dr. Poehlman has agreed to be barred for life from seeking or receiving funding from any federal agency in the future, including all components of the Public Health Service, and to submit numerous letters of retraction and correction to scientific journals related to his scientific misconduct."
It is not clear what compels individuals to perpetrate scientific fraud. Perhaps it is the perceived pressure to continuously achieve at a high level, a willingness to create data to "confirm" the researcher’s own uncorroborated theories, a need to be recognized as a leader and innovator, or a need to compete for funding. This behavior does not equate well to anything outside of the world of research and academe, with the possible exception of athletes taking performance-enhancing drugs. These athletes also are in a highly competitive environment and may feel a need to improve by any means to be "the best" at their sports. They seek the rewards that derive from athletic achievements—the money and fame—and are willing to resort to unsportsmanlike conduct if necessary.
The consequences of cheating in the world of sports may be lifetime bans and derision by fellow athletes and the public. But the conduct of individual athletes seldom affects the athletic community at large. Although the motivations for misconduct in sports and research may be similar, they differ greatly. Scientific misconduct may erode the public trust in the scientific community. It may create a flawed foundation of knowledge on which future research will be built, and it may negatively influence decision makers responsible for creating funding opportunities and public policy. In the case of politically charged issues—stem cell research, for example—scientific misconduct may influence the political process. What is even more disturbing, fraudulent scientific claims can create false hopes among those who have debilitating, even life-threatening diseases.
We—readers, reviewers and editors of scientific journals—need to step up our efforts to prevent and detect scientific fraud. We need to identify consequences sufficient to deter the unscrupulous few whose deeds shadow the accomplishments of all diligent and honest scientists.
