During the late 1970s and the 1980s, an ethics of scientific publication began to evolve. Competition among scientists for academic rewards and research funds, the continued fragmentation and commercialization of science, and reports of scientific misconduct, as well as increasing governmental and legal interference with the inner workings of the scientific community led many within that community to perceive a need for reforms to guide both the conduct of science and the dissemination of scientific information. Journal editors, universities, professional associations, funding agencies, and governments have taken active roles in debating and setting ethical standards and editorial policies for the dissemination of scientific information. In 1978, a self-appointed group of editors, the International Committee of Medical Journal Editors (ICMJE), representing leading general medical journals, met in Vancouver, British Columbia, to set technical guidelines for the submission of manuscripts. These guidelines, the Uniform Requirements for the Submission of Manuscripts to Biomedical Journals, have evolved to include statements for the ethical conduct of authors, editors, and peer reviewers. While the ICMJE statements set international standards for biomedical publishing, the number of journals that adhere to them is unknown (ICMJE, 1991, 1993b). This entry presents an overview of the major ethical issues in biomedical and scientific publishing.
Editorial and Peer Review
The prestige and influence of biomedical journal publication are closely related to the quality control and selection process that precedes publication. Thus, the essential tasks of medical editing are the selection and improvement of articles submitted for publication. These tasks are generally accomplished through processes of editorial review (evaluation by the journal's editorial staff) and peer review (evaluation by experts in a given field who are considered the authors' "peers"). These two processes may overlap, particularly when an editor is also an expert in a manuscript's topic, but editorial review usually focuses on the appropriateness, clarity, and priority of articles for the journal's readership. Peer reviewers are selected by the editor to assess the quality of an article's scientific and technical content and to offer advice about publication. Since decisions regarding rejection, revision, or acceptance are made solely by the editor, the term referee exaggerates a reviewer's advisory role and should be avoided.
Peer review was first used for biomedical publications by the Royal Societies of London and Edinburgh in the eighteenth century, but evolved haphazardly; it was not employed regularly until after World War II (Lock). Two striking aspects of peer review are that it is based almost entirely on uncompensated, voluntary labor and that the peer review system itself has only recently come under scientific scrutiny (Lock; "Guarding the Guardians,"; Rennie and Flanagin, 1994b). Journals follow differing policies about revealing reviewers' identities to authors and authors' identities to reviewers (Lock; "Guarding the Guardians,"; Rennie and Flanagin, 1994b). Some editors believe that disclosure of reviewer identities to authors decreases the potential for bias, while others believe such disclosure leads to less critical reviews. Many biomedical journals do not attempt to remove the identities of authors or their institutions from submitted manuscripts; studies have shown that author identities may be discerned by reviewers from the paper's content or from bibliographic citations, especially in narrow subspecialties (Lock). On the other hand, these same journals do not reveal the identities of peer reviewers to authors. While most editors are impressed by the care and objectivity usually reflected in reviewer comments and recommendations, the anonymous review of papers whose authors are known obviously involves potential for abuse. To maintain integrity in the peer review process, reviewers are expected to disclose any conflicts of interest involved in their review, and editors are expected to be alert to any signs of bias that may interfere with an objective evaluation of the merits of the paper.
Maintaining the confidentiality of an author's work before publication is an important ethical principle in scientific publishing. Most journals inform peer reviewers that the information in unpublished manuscripts is privileged and should be kept confidential, and also require manuscripts to be either returned to the editorial office or destroyed after review. However, maintaining confidentiality depends on an honesty among editors, authors, and reviewers that is nearly impossible to guarantee. Conscious or unconscious intellectual theft by peer reviewers may occur but cannot be measured. Journal editors have a particular responsibility to maintain strict confidentiality about the peer review process, editorial decisions, and all manuscript submissions.
How well do the processes of editorial and peer review work? Many persons involved in publishing recognize the improved quality of articles that have been revised after review, and this has been clearly demonstrated with regard to improvements of study designs and statistical methods ("Guarding the Guardians,"; Rennie and Flanagin, 1994b). Nevertheless, both editorial and peer review are based on human judgments that carry the potential for bias and error.
One form of publication bias is the tendency for papers with statistically significant "positive" results (for example, those showing that a new treatment works better) to be published in favor of papers with statistically nonsignificant "negative" results (for example, those showing that a new treatment does not have any effect or does not work any better than other treatments). Studies have shown that such publication bias exists, but its extent is unknown and controversial ("Guarding the Guardians,"; Rennie and Flanagin, 1992, 1994b). Prepublication bias (the tendency of authors not to submit negative results for publication because the findings are incomplete or nonsignificant or because funding runs out) and postpublication bias (bias in the reception and interpretation of published research data by researchers, funding agencies, editors, and the media) may be more substantial problems. All of these forms of bias can lead to inappropriate medical policies and treatment decisions, especially with new or controversial therapies. Hence, the evaluation of scientific results should be based on their quality and importance, not on their direction.
Despite the fact that university promotion committees evince some shift in the emphasis from the quantity to the quality of publication, academic pressures to publish remain. In many academic circles, achievement is still measured by the length of an individual's bibliography. As a result, authorship of an article published in a peer-reviewed scientific journal carries considerable merit, and consequently, considerable responsibility (Rennie and Flanagin, 1994a).
During the past several decades, the meaning of authorship has become diluted as the number of names appearing in scientific article bylines has grown. Authors have justified lengthy bylines by the increasing specialization of science and the need for collaboration among many subspecialists. But the once-accepted practices of adding the names of a department chair or laboratory chief to the end of bylines (guest authorship), and hiring someone to write up a paper without credit (ghost authorship), have caused many editors to adopt formal policies to curtail inflated bylines (Huth, 1986a, 1986b; Lundberg and Flanagin; Rennie and Flanagin, 1994a) and limit the number of names that can appear in a byline without formal justification.
In 1985, the ICMJE recommended that only those persons who have participated sufficiently to take public responsibility for the work should be authors and that "authorship credit should be based solely on substantial contributions to (a) conception and design, or analysis and interpretation of data; (b) drafting the article or revising it critically for important intellectual content; and (c) final approval of the version to be published" (ICMJE, 1991). Each of these criteria must be met by each person listed in the byline, and the authors must state that they meet these criteria in the cover letter accompanying each submitted manuscript. In the latter half of the 1980s, a number of medical journals, including the Annals of Internal Medicine and the Journal of the American Medical Association (JAMA), began requiring authors to sign authorship statements based on the ICMJE criteria. Anyone who does not meet these conditions but has contributed or assisted significantly can be recognized in an acknowledgment within the article, if he or she has given written permission to be so named (ICMJE, 1993b).
Group authorship results when investigators from many different institutions or participants in study groups, consensus conferences, or working groups prepare reports of their works. Frequently these groups comprise hundreds of investigators, technicians, and specialists. While it is conceivable that each of these individuals contributed critical time and information to the overall work, it is unlikely that each meets the ICMJE authorship criteria. In these cases, those participants who do meet the authorship criteria can be listed with the name of the study group in the byline. Those participants who do not qualify for authorship are then listed in a group box or in an acknowledgment. If all of the participants do meet the criteria for authorship, then the group name can be listed as the sole byline, with the individuals composing the group named in a separate box or the acknowledgment.
Unlike the definition of authorship, there are no established standards for order of authorship, although a number have been proposed, ranging from alphabetical listings to mathematical formulas for determining individual contribution levels and ranking. Many editors agree that authors should be listed according to how much they contributed, with the author who contributed the most listed first and the author who dontributed least listed last (Huth, 1986a, 1986b; Riesenberg and Lundberg). In addition, a number of publications and indexes limit the number of names to be published in a reference list to three, six, or ten. But there is still no consensus on the order of authorship, mostly because there are no widely accepted objective measures of individual coauthors' contribution levels. Editors recommend that authors determine the order of authorship before writing their papers, or before beginning their study, with an agreement to reevaluate the order later if necessary. Editors also recommend that authors solve disagreements over order among themselves, since the authors are in the best position to determine levels of contribution (Riesenberg and Lundberg; ICMJE, 1991).
Another result of the pressures to publish and a driving force behind the need for ethical standards in scientific publication is the practice of duplicate publication. Also known as multiple, dual, or redundant publication, duplicate publication is the simultaneous or subsequent publication of the same article or major parts of an article—methods, results and data, discussion, conclusions, and graphic or illustrative material—in two or more journals or other media, including electronic journals and databases, without notifying the editors (Huth, 1986a, 1986b; ICMJE, 1993b; Iverson et al.). The types of duplicate publication range from selfplagiarism (publishing two or more identical articles or large parts of an article in different journals without citing each article in the texts and references lists) to "salami slicing" (dividing up different parts of the same study for publication in different journals) to sequential publication (reporting follow-up of the same study with additional subjects but without new results). Word-for-word duplication is uncommon, as duplicators usually attempt to alter or disguise the similarities.
Duplicate publication should be distinguished from secondary publication, in which an article or abbreviated version is subsequently republished, in the same or another language, with the consent of both editors. The secondary article should include a footnote on the title page, informing all readers that the information was published previously, and a complete citation to the primary article. Duplicate publication may violate copyright law, and it is unethical for an author to submit duplicate papers to different journals without notifying the editors. By doing so, authors clutter the literature with redundant information; waste the valuable time and resources of editors, reviewers, and readers; and prevent other authors from publishing their work because of limited journal space. To discourage such practices, many scientific journals state in their instructions for authors that they will only consider papers that have not been previously published or submitted to other journals, and some journals will publish notices of duplicate publication, publicly admonishing those authors who publish duplicate articles in violation of the journal's written policies (Iverson et al.).
Conflicts of Interest
Reflecting the increasing commercialization of science and the public doubts about researchers' once hallowed and rarely questioned integrity, financial conflicts of interest are now recognized as another ethical problem for authors and editors. During the 1980s, the public's trust of the scientific community diminished as a result of a number of public scandals and government investigations of biomedical researchers' ties to drugs with potential public health benefits and high financial rewards for stockholders and manufacturers (Relman; Lundberg and Flanagin; U.S. Congress). These cases have generally involved researchers being biased by their direct but undisclosed financial interests, such as stock ownership and paid consultancies. However, there are several other potential sources of author bias: funds from granting agencies, any research or material support, employment, money paid for expert testimony, and honoraria paid for public speaking.
Recognizing that not all financial interests will bias an author, editors disagree over how to handle these financial interests. Most journals publish an author's source of funding or material support, but that is usually because the funding institution requires that it be published. Some journals require authors to disclose all financial interests relevant to the work reported in their submitted manuscripts. If a manuscript is subsequently accepted for publication, the editors of these journals will determine whether it is necessary to publish such financial interests. In this manner, readers can judge for themselves the author's potential for bias from a financial interest just as they can judge an author's potential for intellectual bias based on his or her previously published works or specialty status (Rennie et al.). In 1990, the New England Journal of Medicine instituted a stringent policy prohibiting anyone with relevant financial interests from publishing editorials or review articles in that journal. Critics have argued that such prohibition is scientific censorship.
In 1989, the American Federation for Clinical Research and the Association of American Medical Colleges recommended full disclosure of all relevant financial interests and the possible divestiture of any stock or equity in a company that makes a product the researcher is studying (U.S. Congress). The Editorial Policy Committee of the Council of Biology Editors (CBE) recommends that authors disclose all relevant financial interests to the editors at the time of manuscript submission, and that editors disclose authors' financial interests to reviewers and readers when appropriate (CBE). There is no consensus among editors for the need and extent of such disclosure. In 1993, however, the ICMJE approved a statement that all participants in the peer review and publication process disclose any conflicting interests (ICMJE, 1993a). Some journals with disclosure policies have applied the basic principles of disclosure to everyone in the editorial process, including editors, editorial board members, and in some cases, reviewers (Relman; Rennie et al.).
Fraudulent Publication Resulting from Scientific Misconduct
The publication of a fraudulent article remains the most serious transgression of the ethics of scientific publication. The once generally accepted view that scientific misconduct was rare and committed by a few deviants has been replaced by a view, unsubstantiated, that it is more common and can involve respected scientists from leading institutions. Scientific misconduct has been defined as plagiarism (presenting another's ideas without attribution), fabrication (presenting data or facts that do not exist), falsification (changing or selecting certain data to obtain a desired result, misrepresenting evidence or facts, or misrepresenting authorship), or other serious deviations from accepted practice in the proposing, conducting, or reporting of research (U.S. Department of Health and Human Services). Policy makers have disagreed over the merits of including the phrase deviations from accepted practice in the definition. Some argue that the phrase is too vague and thus open to misinterpretation and overuse (Committee on Science); others argue that it must be included to address misconduct that would not technically be considered plagiarism, fabrication, or falsification. Examples of such deviations include misuse or theft of privileged information by a reviewer or editor, submitting a paper listing several coauthors who are unaware that they are named as coauthors, misrepresenting publication status of articles in a bibliography, or failing to perform funded research while filing reports stating that such work has been done (U.S. Department of Health and Human Services).
Variations in the definition of fraud have caused some confusion, but most editors acknowledge a major difference between fraud and unintentional errors. Although unprofessional and in some cases unethical, the following usually are not considered fraudulent: errors in study design or application of methods, inappropriate use or interpretation of statistics, faulty interpretation or overgeneralization of study results, failure to cite relevant literature or studies, duplicate publication or fragmentary reporting of results, prepublication release of information, publication bias, failure to disclose intellectual or financial conflicts of interests, or violations of experimentation rules protecting humans or animals.
Plagiarism is probably more commonly acknowledged, since it is easier to detect and prove. Detecting and proving falsification or fabrication of data in a published article is not so easy, and it carries grave ethical and legal consequences for editors, authors, institutions, and funding agencies. While an editor has a duty to see that questions of fraud are appropriately and confidentially pursued, the Association of American Medical Colleges, the National Academy of Sciences, and the ICMJE recommend that primary responsibility for investigating cases of suspected fraud rests with the author's institution or funding agency (Association of American Medical Colleges; Committee on Science; ICMJE, 1991). If it is determined that a fraudulent paper has been published, the journal should print—in a timely manner—a retraction, written by the author(s) or an appropriate representative of the institution. Since the validity of any previous work by the author of a fraudulent paper cannot be assumed, the editor must ask the institution to verify the validity of any of the author's articles previously published in the journal or to retract them (ICMJE, 1991).
Protecting Patient Rights
The two major issues regarding patient rights in medical publishing are requirements for the ethical conduct of published research and the protection of patient confidentiality. A now well-established principle followed by all credible medical journals is that reports of experimental investigations of human or animal subjects must include a statement that the research project has been approved by an appropriate institutional review board (IRB). For investigators not covered by a formal ethics review board, the report should state that the researchers have followed the principles of the Declaration of Helsinki (World Medical Association), which includes requirements for freely given informed consent and for the review of the research protocol by a committee independent of the investigator and the sponsor. Many journals also require an additional statement of the manner in which informed consent was obtained from human subjects, since informed consent is a central tenet for ethical research.
Many editors now agree that journal publication should protect patient confidentiality. For example, placing a black bar over the eyes in a facial photograph does not effectively disguise identity. Patients may also be identified from detailed case descriptions. In 1991, the ICMJE published expanded guidelines for the protection of patients' right to anonymity (ICMJE, 1991). These guidelines state that identifying information should be avoided unless it is essential for scientific purposes; informed consent should be obtained for the publication of identifying descriptions or photographs; changing patient data should not be used as a way of securing anonymity; and journals should publish editorial policies to preserve patient confidentiality (ICMJE, 1991).
One problematic area regarding patient anonymity is the publication of pedigrees from genetics research, since the family as a whole or individual family members can sometimes be identified from pedigree information. Following the ICMJE guidelines, identifying information should be deleted if possible, but pedigree data should not be altered. Pedigree publication is complicated by the fact that a large number of family members may be involved, not all of whom may have given consent for, or even be aware of, the collection of family data. A requirement for informed consent for publication from each individual member of a large pedigree may be impossible to meet, particularly if family members disagree about publication. Whether some kind of group consent would be ethically permissible, or whether identifiable pedigrees should not be published without the consent of each individual family member, remains an unsettled issue.
Release of Information
Scientific journals play a major role in informing the public, as well as health professionals, about biomedical developments. This function involves a balance between the timely release of information and the adequate evaluation of the quality of the information. Conflicts sometimes occur between scientists, who want rapid dissemination of new or controversial research findings; editors, who as gatekeepers want to make sure that only accurate and valid scientific information is released; and the news media, which compete with each other to be the first to publicize new scientific information. The process of scientific publication after peer review takes time. Some investigators have chosen to short-circuit this traditional process by announcing results at a news conference rather than waiting for a paper to be evaluated by a scientific journal. Advocates for a particular disease (acquired immunodeficiency syndrome [AIDS], for example) have also pressed for faster release of research results. Even if well-intended, such attempts to bypass careful evaluation and publication may result in the dissemination of misinformation (Angell and Kassirer).
In 1969, Franz Ingelfinger, then editor of the New England Journal of Medicine, promulgated a policy (subsequently known as the Ingelfinger rule) that manuscripts would be considered for publication only if their substance had not been submitted or reported elsewhere. Other journals adopted similar policies to discourage both duplicate publication and the public dissemination of results before peer review and publication. Such policies have been criticized as self-serving on the part of journals, but they usually exempt presentations at scientific meetings (including published abstracts and media coverage from such meetings) and the rare situations when an appropriate public health authority determines that there is an immediate need for dissemination. Some medical journals also ask news media to observe a press embargo for a brief period to allow physician subscribers to read and evaluate information before their patients begin seeing it in the media.
Copyright protection covers text and illustrative material—whether in print or electronic (digital) format. U.S. copyright law provides that the creator of a written work, the author, owns all legal rights to that work for his or her life span plus fifty years, unless the author transfers those rights to another party. Two exceptions to individual copyright ownership are works prepared by employees of the U.S. government and works made for hire, in which an individual, either by an employment mandate or by contract, agrees in writing that all work prepared within the scope of employment or contract is the property of the employer or contractor (Copyright Law of the United States of America). Different countries have different copyright laws, but the Universal Copyright and Berne Conventions protect works published and distributed in other countries.
Most journals require authors to transfer copyright to their publishers before publication, giving the publisher exclusive rights to the work after publication. Therefore, anyone who wishes to reprint or adapt from an article (in part or whole) must receive written permission to do so from the publisher. However, certain uses of a published work without permission from the owner—such as photocopying for teaching, scholarship, or research purposes—may not be an infringement of copyright under the provisions of "fair use." Fair use can be difficult to justify in court and must take into account the following factors: (1) the purpose of the use, including whether it is educational or commercial;(2) the nature of the copyrighted work; (3) the amount of the copyrighted work to be used; and (4) the effect of use on the potential marketability or value of the copyrighted work (Copyright Law of the United States of America).
Rights to Unpublished Data
Unlike rights to copyrighted work, rights to unpublished data are difficult to define, and most ethical dilemmas concern access to rather than ownership of such information. Unpublished scientific data include written and electronic laboratory notes, experimental materials, project records and observations, databases, descriptions of methods and processes, analyses, and illustrative material. Traditionally, unpublished scientific data have been owned by their creators—the scientific investigators—and most scientists believe they have a duty to share data with their peers and, when appropriate, with the public. Any data reported in a published article become the property of the publisher, but rights to relevant, supportive data not reported in a published article (sometimes called raw data) are not transferred to the publisher. Problems arise when investigators, institutions, the government, and the public compete for control of and access to the same data. For example, who should have first rights to publication of research data: the principal investigator, the coinvestigators, or the institution that funded the research? Legally, the investigator controls access to unpublished data, except under the following circumstances: (1) the investigator is an employee of an organization that claims rights to any work conducted by its employees; (2) the investigator is under federal contract or has received a federal grant to perform the work; or (3) a court decides that public interest in the data outweighs the interest of the owner (CBE). Government or industrial sponsorship of research may impose specific restrictions on data control and sharing, particularly when such data are proprietary or commercial. This area of law will continue to evolve as electronic technology makes data ownership and access more difficult to define and control by narrow standards and laws.
While it is generally agreed that data must be kept in an accessible format for a reasonable period of time, no standard has been universally accepted, because different types of data from different specialties require various modes and spaces for storage, which can be prohibitively expensive. Some institutions have recommended three or five years, and longer periods for data that support publications (Committee on Science). The National Research Council Committee on National Statistics recommends and many journals require that editors have access to data during the peer review process, which means that the data must be maintained until publication (CBE). Some journals require authors to provide data to editors for their evaluation if requested, but this requirement does not have a time limit. Some journals require authors to send their data to national or international storage centers at the time of publication.
Disputes over who has rights to use scientific data have caused ethical dilemmas for editors. For example, what should an editor do with a manuscript from an author that reports an analysis of unpublished data originally collected and analyzed by another author? The ICMJE and the Committee on National Statistics recommend that editors consider such secondary analyses on their scientific merit as long as full credit and appropriate citations are given to the original data collections (ICMJE, 1991; CBE). Other open questions concern the nature of sharing data, which is a vital part of the scientific enterprise. Should there be restrictions on the access, use, and citation of unpublished works by other authors and investigators? Most scientists and editors would argue that such restrictions would stifle scientific exchange. But what about access to unpublished data by those outside the scientific community, such as representatives of the media, the courts, and people with commercial interests? Many of these questions are currently under debate, and whether or not access will be widened or restricted is difficult to predict.
Advertisements for pharmaceutical products and medical/laboratory devices provide major financial support for biomedical publications. Advertising income is essential for many large biomedical publications since their costs would not be met by subscription revenue. Whether this situation represents one aspect of the success of the free enterprise system or a major ethical problem for editors is a matter of controversy.
To protect a journal's integrity and credibility, complete separation between advertising and editorial decisions is essential, and advertisers should have no influence on editorial content. Advertisements, including advertorials, should have a distinct appearance or labeling so that readers can readily distinguish them from editorial content, and ads for a product should not be placed adjacent to editorial material dealing with the product or disorders for which it might be used (Rennie). Publication of industry-sponsored journal supplements is problematic, since the supplement's editorial content may be selected or influenced by the sponsor to favor their products, and the review process may not be as rigorous or as independent as it is for the journal's regularly published issues.
The accuracy of advertisements in medical publications is more controversial. The purpose of advertisements is promotional, and studies have shown that the prescribing behavior of physicians is indeed influenced by advertisements. Because of their effect on the health of the public, advertisements for drugs and medical devices are regulated by a government health agency in many countries. In the United States, this responsibility lies with the Food and Drug Administration (FDA), which reviews and approves marketing and labeling (the package insert that describes the indications and side effects of a drug) but does not routinely review or approve advertisements prior to their dissemination. However, the FDA does review advertisements after publication and can require companies to withdraw or publicly correct ads that it determines to be inaccurate or misleading.
The standards by which print advertisements should be judged and the method of enforcing standards remain unsettled. Some have recommended the development of multidisciplinary review boards, such as the Canadian Pharmaceutical Advertising Advisory Board, to review and approve medical advertisements before their dissemination.
Enforcement of Ethical Standards
The enforcement of ethical standards in scientific publishing is a responsibility shared among authors, institutions, funding organizations, peer reviewers, and editors. Authors are primarily responsible for upholding the scientific commitment to a search for truth, accepting responsibility and credit for the work that bears their names, and fully disclosing any conflicts of interest. Institutions where research is performed and organizations that fund research share the main responsibility for ensuring that studies are designed and conducted ethically, and also for investigating and sanctioning allegations of misconduct. Peer reviewers are charged with performing objective and timely appraisals of papers submitted for publication, while maintaining strict confidentiality and disclosing their own conflicts of interest. Editors should exercise sound judgment and objectivity in selecting papers for publication, maintaining vigilance for any ethical problems, and ensuring that authors, reviewers, and institutions fulfill their responsibilities. Clear ethical standards and implementation policies are certainly desirable, and editors have taken the lead in setting standards and policies (U.S. Congress). Yet the ethics of scientific publication is based on trust, and obsessive "policing" of the research community and the publication enterprise could be counterproductive. Persistent emphasis on the importance of maintaining ethical standards in the entire research process, from initial research ideas to their eventual publication, should be an expectation shared by all involved in that process. However, defining and enforcing such standards will be an even greater challenge as the electronic revolution extends the traditional boundaries of authorship and scientific publication.
richard m. glass
annette flanagin (1995)
Angell, Marcia, and Kassirer, Jerome P. 1991. "The Ingelfinger Rule Revisited." New England Journal of Medicine 325(19): 1371–1373.
Association of American Medical Colleges. Ad Hoc Committee on Misconduct and Conflict of Interest in Research. 1992. Beyond the "Framework": Institutional Considerations in Managing Allegations of Misconduct in Research. Washington, D.C.: Author.
Committee on Science, Engineering, and Public Policy. Panel on Scientific Responsibility and the Conduct of Research. 1992–. Responsible Science: Ensuring the Integrity of the Research Process, vol. I. Washington, D.C.: National Academy Press.
Copyright Law of the United States of America, as Contained in Title 17 of the United States Codes. 1993. Washington, D.C.:U.S. Government Printing Office.
Council of Biology Editors (CBE) Editorial Policy Committee. 1990. Ethics and Policy in Scientific Publication. Bethesda, MD: Author.
DeAngelis, Catherine D.; Fontanarosa, Phil B.; and Flanagin, Annette. 2001. "Reporting Financial Conflicts of Interest and Relationships between Investigators and Research Sponsors." Journal of the American Medical Association 286(1): 89–91.
Flanagin, Annette. 2000. "Human Rights in the Biomedical Literature: the Social Responsibility of Medical Journals." Journal of the American Medical Association 284(5): 618–619.
Fontanarosa, Phil B., and Flanagin, Annette. 2000. "Prepublication Release of Medical Research."Journal of the American Medical Association 284(22): 2927–2929.
Fontanarosa, Phil B.; Flanagin, Annette; and DeAngelis, Catherine D. 2000. "The Journal's Policy Regarding Release of Information to the Public." Journal of the American Medical Association 284(22): 2929–2931.
Garrett, Jinnie M., and Bird, Stephanie J. 2000. "Ethical Issues in Communicating Science."Science and Engineering Ethics 6(4): 435–442.
"Guarding the Guardians: Research on Editorial Peer Review: Selected Proceedings from the First International Congress on Peer Review in Biomedical Publication." 1990. Journal of the American Medical Association 263(10): 1317–1441.
Huth, Edward J. 1986a. "Guidelines on Authorship of Medical Papers." Annals of Internal Medicine 104(2): 269–274.
Huth, Edward J. 1986b. "Irresponsible Authorship and Wasteful Publication." Annals of Internal Medicine 104(2): 257–259.
International Committee of Medical Journal Editors (ICMJE). 1991. "Statements from the International Committee of Medical Journal Editors." Journal of the American Medical Association 265(20): 2697–2698.
International Committee of Medical Journal Editors (ICMJE). 1993a. "Conflicts of Interest." Annals of Internal Medicine 118(8): 646–647.
International Committee of Medical Journal Editors (ICMJE). 1993b. "Uniform Requirements for Manuscripts Submitted to Biomedical Journals." Journal of the American Medical Association 269(17): 2282–2286.
International Congress on Peer Review in Biomedical Publication. 1991. Peer Review in Scientific Publishing. Chicago: Council of Biology Editors.
Iverson, Cheryl; Dan, Bruce B.; Glitman, Paula; et al. 1989. American Medical Association Manual of Style, 8th edition. Baltimore: Williams & Wilkins.
Jones, Anne Hudson, and McLellan, Faith, eds. 2000. Ethical Issues in Biomedical Publication..Baltimore: Johns Hopkins University Press.
Lock, Stephen. 1986. A Difficult Balance: Editorial Peer Review in Medicine. Philadelphia: ISI Press.
Lundberg, George D., and Flanagin, Annette. 1989. "New Requirements for Authors: Signed Statements of Authorship Responsibility and Financial Disclosure." Journal of the American Medical Association 262(14): 2003–2004.
Olson, Carin M.; Rennie, Drummond; Cook, Deborah; et al. 2002. "Publication Bias in Editorial Decision Making." Journal of the American Medical Association 287(21): 2825–2828.
Relman, Arnold S. 1984. "Dealing with Conflicts of Interest." New England Journal of Medicine 310(18): 1182–1183.
Rennie, Drummond. 1991. "Editors and Advertisements: What Responsibility Do Editors Have for the Advertisements in Their Journals?" Journal of the American Medical Association 265(18): 2394–2396.
Rennie, Drummond, and Flanagin, Annette. 1992. "Publication Bias: The Triumph of Hope Over Experience." Journal of the American Medical Association 267(3): 411–412.
Rennie, Drummond, and Flanagin, Annette. 1994a. "Author-ship! Authorship! Guests, Ghosts, Grafters, and the Two-Sided Coin." Journal of the American Medical Association 271(6): 469–471.
Rennie, Drummond, and Flanagin, Annette. 1994b. "Selected Proceedings from the Second Internatinal Congress on Peer Review in Biomedical Publication." Journal of the American Medical Association 272(2): 91–173.
Rennie, Drummond; Flanagin, Annette; and Glass, Richard M. 1991. "Conflicts of Interest in the Publication of Science." Journal of the American Medical Association 266(2): 266–267.
Riesenberg, Don, and Lundberg, George D. 1990. "The Order of Authorship: Who's on First?" Journal of the American Medical Association 264(14): 1857.
Smith, Richard. 2001. "Maintaining the Integrity of the Scientific Record."British Medical Journal 323(7313): 588.
U.S. Congress. House Committee on Government Operations.1990. Are Scientific Misconduct and Conflicts of Interest Hazardous to Our Health? Washington, D.C.: U.S. Government Printing Office.
U.S. Department of Health and Human Services. Office of the Assistant Secretary for Health. 1991. First Annual Report: Scientific Misconduct Investigations: Reviewed by the Office of Scientific Integrity Review, March 1989-December 1990. Washington, D.C.: Author.
World Medical Association. 1990. "Declaration of Helsinki." Bulletin of the Pan American Health Organization 24(4): 606–609.