CRS-23 does not qualify one to be listed as a coauthor. Additional contributions of lesser significance would be acknowledged in another part of the paper. The American Association for the Advancement of Science-American Bar Association National Conference of Lawyers and Scientists addressed the issue of multiple authors. It was noted that: As a general rule, every named author of a publication The Institute of Medicine's Committee for the Study of Responsible Conduct of Research has also proposed that journal editors make it mandatory that authors of submitted work sign a statement attesting to their level of participation and contribution to the research. In addition, co-authors would also be required to sign statements accepting their responsibility for the accuracy and validity of the data in the report.39 Arnold Relman, New England Journal of Medicine, supports a proposal of having various levels of authorship.40 40 The primary author would be the one who actually designed the study and generated and analyzed the data. Secondary author/s would be restricted to those who assisted, collaborated, or provided moderate consultation. In addition, all authors would be required to sign statements certifying their level of involvement in the project. This proposal has been termed impractical by many scientists because of the interdisciplinary nature of research in many of the laboratories and institutions. The American Chemical Society (ACS) echoes the concerns of other scientific associations by establishing ethical guidelines for publication of its chemical research for editors, authors, and manuscript reviewers. In issuing the guidelines, the ACS stated that: 38 Project on Scientific Fraud and Misconduct. Report on Workshop Number One. American Association for the Advancement of Science-American Bar Association National Conference of Lawyers and Scientists. Washington, 1988. p. 120. 39 National Academy of Science, Institute of Medicine. The Responsible Conduct of Research in the Health Sciences. Washington, 1989. p. 38. 40 Coughlin, Ellen K., p. A6. CRS-24 These guidelines are offered not in the sense that there is According to the ACS guidelines, it is axiomatic that editors publish retractions or errata of any research proven to contain fabrications or falsifications.42 While several groups and associations (including the ICMJE, the Institute of Medicine, the National Institutes of Health, the Association of Medical Colleges, the Association of American Universities, and the Endocrine Society) have invested considerable time in establishing guidelines to address allegations of misconduct in published or submitted material, others still face substantive procedural difficulties and continue to debate policies and procedures. Paul J. Friedman, School of Medicine, University of California, San Diego, acknowledges that some biomedical journals have implemented the policies established by the ICMJE, while others have failed to even publish retractions or erratum of a study offered by a host institution. In 1987, after a review of Robert Slusky's bibliography of 137 articles, the administrators at the University of California, San Diego (UCSD) found 12 articles to be fraudulent and another 48 questionable. UCSD administrators notified journal editors of the results of their review and indicated that it was the journals' responsibilities to publish retractions. Reportedly, because of the threat of legal liability from third party retractions, some journals did not publish retractions or errata for the invalid work. In other instances, journal editors have allowed researchers to offer corrections to a study by publishing a new paper that replaces the previously flawed study. Other editors have debated whether to withdraw a study from publication if requested by the principal author over the objections of the coauthor. 43 It appears that editors find it is easier to handle cases of falsified or plagiarized research after it occurs, rather than prevent its publication in the first place. Their overarching concern is that of protecting the journal from threats of libel. Rosemary Chalk has stated that: 41 American Chemical Society. Ethical Guidelines to Publication of Chemical Research, Jan. 1985. 42 Zurer, Pamela S. Misconduct in Research. It May be More Widespread than Chemists Think. Chemical and Engineering News, v. 65, Apr. 13, 1987. p. 11. 43 Dalton, Rex. Journals Slow to Retract Slutsky Research Errors. The Scientist, v. 2, Mar. 7, 1988. p. 1. CRS-25 Publishing accusations which question the integrity of a EVALUATION OF THE SCIENCE COMMUNITY'S The science community is taking positive steps in each area of intervention-prevention, detection, and investigation-with regard to the issue of misconduct in science. In some cases, the action is still at the stage of consideration or recommendation. So, an issue facing policymakers concerns the creation of incentives to motivate the appropriate parties to adopt and implement the recommendations. Another policy issue concerns the effectiveness and efficiency of the implemented recommendations. Some critics assert that a few of the recommendations even when implemented would be inadequate to address the dimension of the problem they target. These critics have offered either additional or alternative solutions. Table 5 lists the recommendations/actions of the science community as well as the additions and alternatives suggested by critics. (See page 25.) Options for Preventing Misconduct As table 5 indicates, a number of steps have been taken to prevent misconduct (or to insure quality and integrity) in research. Among the various options there is at least one action or recommendation which addresses each of the possible factors contributing to the occurrence of scientific misconduct, i.e., competition, financial conflict of interest, increasing lab size, and co-authorship. Each of these options has been well received or at least seriously considered by many of the interested parties. However, there are other options which have been considered by at least a few individuals. Two of these options are extending grant periods and creating/implementing stiffer sanctions. Extension of Grant Periods In FY88, the average life of a new grant was 3.25 years, with approximately 28 percent of the new grants for 1 year, 25 percent for 3 years and 38 percent for 5 years." Some individuals suggest extending grant 44 Project on Scientific Fraud and Misconduct, p. 28. 46 Moore, Robert. Division of Research Grants, NIH. Personal communication, May 19, 1989. CRS-27 periods as an additional strategy to address the problem of competition in science. Generally, researchers are rewarded--promoted, given tenure and funds, and receive acclaim-on the basis of results. Such benefits are less available if they fail to either prove or disprove something. By extending the grant period, researchers are given more time to achieve results. Hence, whatever pressure exists to motivate some to cheat, e.g., to fabricate data, would be reduced. While extending the grant period may have the positive effect of reducing the pressure to cheat, it can have negative effects as well. For example, some suggest that it can limit a funding agency's ability to respond to new issues or crises by tying up significant sums of money in long term grants. Also, the congressional appropriations process puts constraints on how far grant life can be extended. Application of Criminal Sanctions The PHS guidelines and NSF regulations for dealing with possible misconduct in science list a number of sanctions which are classified according to degree of severity. Among the most severe sanctions are the immediate termination of any active awards and debarring a researcher, for a specific period of time, making him ineligible for receiving Federal funds. In addition, any of the sanctions may involve the recovery of funds. The guidelines and regulations do not specify criminal sanctions. But such sanctions do exist. According to Harold Edgar, Professor of Law at Columbia University, anyone who knowingly and willfully falsifies, conceals or If a researcher is found guilty of lying to the Government in grant applications or in progress reports, he may be fined $10,000 or imprisoned for 5 years. To date, only one researcher has been criminally prosecuted for making false statements on grant applications-Stephen Breuning, a former University of Pittsburgh researcher, pleaded guilty to charges of fabricating data on the effects of tranquilizers on retarded children. According to some experts, the 16 Wheeler, David. Experts Weigh Federal Prosecution of Scholars Who Fabricate Results. The Chronicle of Higher Education, Mar. 1, 1989. p. A5, A7. |