Click for next page ( 18


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 17
4 ANALYSIS ~ meetings before and after he Sir work~hcp~ the commit arrived at assumptions and fires that well] form the basis for the evaluation of polity prc~;a:ts arx] final rations acntair~ In this report. In the ad ence of data d ~ nting the in Angrily of existing research practice= and the level of m~soonduct in hearth sciences research, the committee relied on expert opinion. m ese assumptions and findings may not be share] by all members of the research community ~ indent, they were not share] by all workshop participants but they quickly emerged as a consensus of the committee. These assumptions and findings deserve expla ~ tion and clarification because they form the foundation for this report. AXONS AD EIND~ o Scientists should develop and maintain quality and accuracy in research practice by self-regulation, extensive reliance on each investigator's professional st end arise and the traditions and collegiality that characterize research institutions. The balance between emphasizing individual and emphasizing oollegial responsibility for the integrity of research may vary by discipline, research field, or institution, but self-regulation is the principal system used by the government and the research community to provide integrity and quality in research. m e nature of the scientific enterprise is such that only those persons familiar with the methods and analytical basis of science are capable of understanding and evaluating the research results of others. Scientists are not separate freon society' however' and they are subject to the same laws and standards that govern responsible social behavior. m e norms of responsible research practice are usually ccmmunicate~d through oral tradition and personal example. m ese norms have exercised a significant but poorly understood role in regulating the conduct of individual investigators and in encouraging honesty in academic research. Self-regulation includes peer review and replication of reported research findings. Peer review evaluates the adequacy of scientific methods, the significance of the research, and the consistency of the reported findings. Replication involved repeating the work of others. It may confirm or contradict earlier work, a situation that requires explanation. 17

OCR for page 17
The argument that replication and peer review will guarantee the integrity of all research is incorrect. Many studier are not replicated in detail (it would be highly inefficient if then were). Trust in . researchers, descriptions at- their methods and findings has been fundamental to scientific communication. In the past, scientific skepticism has not extended to the honesty of an investigator's factual statements, but has been directed toward interpretation of reported results. ~ ghost the discussions, the committee agreed that the primary responsibility for the oversight of quality and responsible research practices in the health sciences Should continue to rest with the research community itself. At the same time, the committee recognized limitations of the existing self-regulatory system in encouraging responsible research con ~ at and detecting scientific misconduct. O A variety of informal and formal practice= and procedures currently exist in the academic research environment to assure and maintain the high quality of research conduct. Until very recently, governmental and private institutions a that the principle of self-regulation was sufficient to promote and maintain the integrity and quality of federally funded research. Recent cases of scientific fraud, however, have challenged the wisdom of relying entirely upon investigator honesty and unwritten collegian standards as the sole means of assuring integrity and quality in research. These cases have fostered a perception that existing institutional and professional controls have been inadequate to assure Integrity in federally funded research. The effectiveness of ~elf-regulation depends upon those practices and procedures that have evolved over time as part of the scientific method. Such practices are intrinsic to the conduct of science and they are expected to be a fundamental part of the professional training of each scientific investigator. Informal practices include collegian interaction, skepticism, and criticism; sharing of research ~ ta, methods, and materials; serving as mentors and role models; consensual assignment of authorship crept; and post-publication review by colleagues. Formal mechanisms that promote responsible research conduct include the peer review systems administered by scientific journals and funding agencies; departmental research seminars; the criteria, policies, and procedures that govern appointment and promotion decisions; the use of "visiting comic" and interdepartmental reviews in evaluating the quality of graduate research and training pro grams; and the coordinating committees and data-collection audits used in mwiticenter clinical research programs. However, the use of these mechanisms varies widely among different academic institutions. 18

OCR for page 17
Die committee believes that existing Ins trysts of peer review and university oversight in academic research are f~ntal ly saux1, but it recognizes that many of these pi are not uniformly followed and investigator practic ~ vary. M ~ of the existing institutional controls rely upon the effectiveness of social and professional sanctions. It appears that these controls may have limited impact on some investigators. O Few academic institutions have established explicit standards for responsible research practices, such as guidelines for the recording and retention of research data or for inclusion as an author. As a first step in promoting scientific responsibility and providing quality assurance, some research institutions and scientific organizations are trying to enhance the visibility and effectiveness of generally accepted research practices already in place. In adaption to institutional guidelines, the conduct of research is governed in some matters by various professional and regulatory standards. For example' federal and state Jaws govern the use of animals at human subjects in research; the use of hazardous research equipment arxt materials; He use of r~inant-lX~ maniac; and general laboratory safety. Professional codes of ethics developed by individual scientific societies ~ so establish certain standards for the rx3uct of Joseph. For the Apt part, however, p~vfemcion~ and rotatory stanzas are root ~n~ ~ an explicit Or into the social setting and training of scientific irIvestigators. He committee has recei~rec] reports that same institutional investigations of allegations of scientific Truscor~ct are h ~ by the absence of written standards against which to measure charges of improper practice. For example, same investigators fail to keep research data and are unable to document or verify their results when questioned. Others may seriously neglect their supervision and training responsibilities or may request "honorary authorships' on publications for which they made no significant contribution. The comity believes Hat the absence of explicit institutional scarabs allows the r~ar~h system to tolerate subs activities by a small Or of individual i~r~:tigators Ho fail to Serve generally accepted practices. EDrthermore, the absence of a Satanism to enforce these stanzas= leads to a perception ~ at the institution or the profession is unwilling or unable to correct abusive practices. O There are very few courses of instruction dedicated to communicating professional standards and the ethics of research practice to young scientists. Although some university courses and publications offer instruction in the standards and ethics of research practice, most science students do not receive formal exposure to this topic. me absence of these 19

OCR for page 17
courses limits the cpportuni~ for discussion of the norms arm ideals that shalld guide resort practice and for identifying amount or deviant practices. The lack of instruction In this area also creates a perception that the faculty places refire Isis on Me results than the methods of scientific research. The communication of the ideals of science, its values and methods, traditionally occurred through indivi~r~H discussions between senior investigators and students. Given the increased size, complexity, and heterogeneity of the research training process, the committee believes that reliance on these discussions alone is not sufficient to provide effective instruments of professionalization and education. traditions and styles of governance that assume professional integrity and that place great value on the independence and collegiality of individual faculty. O m e culture of the American university is distinguished by In contrast to some other academic systems, American universities are not hierarchical structures. m err demonstrated effectiveness in generating new knowledge and training young scientists depends greatly upon investigator autonomy and the collegial good will, trust, and academic freedom that are the foundations of academic governance In our society. Although the committee agrees that American biomedical research and training can benefit fern improvements In policies and practices, it believes that some administrative changes cculd create more serious problems than those they are designed to alleviate. Strict or intrusive regulations governing the details of research conduct may damage the creativity and quality of the research environment, especially if those regulations are unsupported by documented evidence of widespread dishonesty or-ethical misconduct by investigators. m e collegiality of the faculty also discourages individuals from revealing negative information about a colleague or student. If faculty are to be encouraged to take a more active role in exposing the deficiencies and weaknesses of their peers, they must be assured that their institutions are prepared to act on their information and that whistle-blowers will not be punished for exposing substandard or deviant practices. The committee believes that better education of academic officials, faculty, and students and, ", there appropriate, legislative protection could help reduce the barriers that discourage the reporting of scientific misconduct when it occurs. O Investigations of a small number of publicly reported cases of scientific fraud and other research misconduct suggest that a mix of factors contributed to this deviant behavior. One of these was an 20

OCR for page 17
unhealthy research environment that failed to discourage (or even tolerated) sloppy or careless research standards. Cases of serious m~soon~uct in science are rare events and there are only a handful of studies of their origins and rouses. Some factors that have keen identified ~ these preliminary analyses include denial, self-decepkion, and unwillingness to believe that such actions could be detected and punished. Some of those found guilty of fabricating or distorting research results denied that their practices were wrong or substandard, often using the expression that "everyone does it." to justify their behavior. The committee believes that, in the long run, the quality of the research environment may be more damaged by sloppy or careless research practices and apathy than by incidents of research fraud or other serious scientific misconduct. The committee did not have the resources to try to conduct a systematic study of the integrity an] quality of current research practices or incidence of scientific fraud in the health sciences. But preliminary studies and workshop discussions suggest that the research community tolerate.= too many substandard practices. These abuses must be corrected to restore a sense of moral integrity and professionalism in research. there are several kinds of contemporary research activities that Lie on Me edges of acceptable professional practice; many researchers would regard them as irresponsible practices. Examples Include abuses of multiple authorship, such as The practice of "gift authorship"; repetitive publication of short-term research results; the neglect of young researchers by their peers and mentors; inadequate training and supervisory practicer; and sequestering or withholding of research data from peers and colleagues. Although the committee believes that serious misconduct in science is rare and is ultimately a manifestation of individual deviance, it con crud-= that institutions fail to detect and correct early deviant behavior, primarily because of an excessively permissive research environment that tolerates sloppy and careless practices. the committee also believes that substandard pr-acti~= are enccuraged by funding pressures and an overemphasis on publication as the main means of achieving status and recognition for scientific advancement and research support. o Increasing budgetary and competitive pressures in science demand that local research institutions and government research funders develop standards to ensure responsible research practices to ensure the integrity of the academic research enterprise. m e committee identified a nope for higher professional stand ard5 at all levels in the research system. m ere was consensus that, although the fundamental values and standards of the research community are appropriate, the expression and implementation of these standards 21

OCR for page 17
are insufficient to promote responsible research practices in an increasingly large, heterogenec us, and competitive research environment. New and comprehensive guidelines should be developed by the research community to clarify traditional practices, to strengthen the mix of formal policies and informal practices currently In place, and to correct actions that seriously deviate from these standards. The committed believes that the most effective sites for the develcpment of these guidelines are the immediate social setting of the investigators, represented by the laboratory and clinical research centers, universities, professional societies, and scientific journals. The adoption of such standards will require leadership from government and other funding sources. In developing final recommendations, the committee sought to define appropriate roles for government agencies and the Congress= that would stimulate local institutional and professional efforts without creating an unjustifiable regulatory burden on the research community. O Effective institutional reforms to improve integrity and responsible research practices require better understanding of the key factors that influence professional development and performance in science. During the workshop, many participants identified a need for additional research to clarify the basic factors that influence professional conduct in the contemporary research environment. In particular, they cited the need to evaluate the experiences of different kinds of research institutions In training investigators and monitoring, rew;o~ling, or correcting research practices. Other participants urged that extensive studies of the incidence and prevalence of scientific misconduct should be undertaken before the development of changes In resear1 h oversight. substantive Institutional policy reforms ideally should be based on a deeper understanding of professional behavior and the effects of training, role models, peer pressure, and the structure of the research system on professional development and behavior. However, the limitations of current understanding of these processes should not delay the development of explicit guidelines when there is clear consensus about what represents responsible research and training practices. 22