Journal Science to tighten review procedures
The leading journal Science has been advised to modify its review procedures to identify and weed out papers based on fraudulent research results, in a report on the publication in the journal of papers by the now discredited South Korean stem cell scientist Woo Suk Hwang. In March 2004, Science published a paper by Dr Hwang and his colleagues which apparently described the development, from a cloned human blastocyst, of human embryonic stem cells with the potential to become any cell in the body. Just over a year later, in June 2005, a second paper was published in which Dr Hwang and his team claimed to have isolated the first human embryonic stem cell lines, specifically tailored to match the DNA of patients suffering from disease or spinal cord injury. Science subsequently retracted both papers in January 2006, following the revelations that Dr Hwang's 'findings' were the result of serious research misconduct. The journal's editorial board set up a committee of leading scientific journal editors and stem cell scientists to investigate its handling of the fraudulent papers, and the committee's report, along with the Editor of Science's response, have now been published on the Science website. According to the report, the committee found that the reviewers of the fraudulent papers had adhered to Science's procedures and 'made a serious effort - substantially greater than that for most papers published in Science - to ensure that the science was sound'. However, they note that current procedures are based on an assumption of trust in the basic integrity of the vast majority of researchers. 'In fact, [Science] now receives a small number of papers that are either intentionally misleading...or substantially distorted by self-interest,' the report's authors write. They believe that the cachet of having an article published in leading journals such as Science or Nature offers an incentive for some scientists not to follow the rules. 'This problem has a significant impact on all of science, since trust in the system is essential, and since Science and Nature are seen to speak for the best in science,' the report states. The committee notes that the number of papers submitted to science means that increasing the level of scrutiny for all of them would be simply unfeasible. However, it recommends that for papers which are likely to be highly visible or influential, the journal should institutionalise a 'healthy level of concern'. 'It is essential to develop a process by which papers that have the likelihood of attracting attention are examined particularly closely for errors, misrepresentation, deception, or outright fraud,' the committee writes. They predict that in the near future, such papers could come from the fields of climate, energy, health and nanotechnology. They advise the journal to carry out a 'risk assessment' for all papers accepted for publication. This new procedure would ask questions about the probability of the work being intentionally deceptive or simply wrong, and the consequences of this for the reputation of Science and science, as well as issues such as public policy, intellectual property or academic credit. The report also recommends the establishment of a method to clarify the contributions and responsibilities of authors and co-authors, and calls for more extensive information to be put into the published-supporting material. 'Primary data are essential and should be available to reviewers and readers,' the authors write. Finally, they advise Science to work with other high profile journals such as Nature to establish common standards. 'It would be undesirable to have authors choose a journal for submission based on standards, or the lack of standards, of the type discussed here,' the report cautions. Responding to the report, Science's Editor-in-Chief, Donald Kennedy, writes: 'We are committed to accepting the major findings of the Report, and to making our new procedures clear to authors, reviewers and readers as they are developed.' He agrees with the report's authors that the publication of the papers was a 'bad outcome.' 'By making the recommendations and our response public, we intend to emphasise our commitment to safeguard the review process from abuse by intentional misconduct,' he writes, adding that readers of Science could expect to see revisions to the guidelines for contributors in the near future. 'We will continue to consult broadly with the scientific community as we consider other changes in response to the Report,' he concludes.
Countries
South Korea, United States