Guidelines for the Responsible Conduct of Research at Yale University School of Medicine
Endorsed by the Board of Permanent Officers, February 13, 1991 and on April 23, 1993, to be used as a working document for educational purposes and for setting general standards of responsibility within the School of Medicine. Revised 9/97.
Federal and private granting agencies, Congressional oversight committees, and national organizations including the Association of American Universities, the Association of American Medical Colleges, and the Institute of Medicine, among others, have all perceived the need for universities and other research intensive institutions to adopt guidelines which clarify each institution's expectations about the professional standards to be observed in the conduct of scientific inquiry.
In 1982, the University published a policy statement on collaborative research (Weekly Bulletin and Calendar, Sept 13-20, 1982) which condemned academic fraud in all of its forms and which assigned to each collaborating investigator equal responsibility for the quality and integrity of a publication resulting from collaborative research. Subsequently, the University published a set of specific policies and procedures for dealing with allegations of academic fraud (Weekly Bulletin and Calendar Oct 30-Nov 6, 1989). The present guidelines are concerned with principles and practices which focus on ensuring the highest possible standards for conducting biomedical research.
Investigations of specific instances of academic fraud have repeatedly identified institutional pressures and deficiencies which, although not directly responsible for misconduct, nevertheless create an atmosphere conducive to unacceptable scholarly conduct. These factors include perceived pressures to publish large numbers of research papers, excessive emphasis on competition and secrecy in scientific inquiry, inadequate attention to systematic collection and retention of research data, insufficient supervision of relatively inexperienced investigators, and inadequate interaction between investigators participating in collaborative research. The following guidelines are intended to address these factors, but they can serve only as general indications of expected standards of professional conduct, not as rigid rules. Scholarly activities are too complex and varied to permit universally applicable, absolute directives.
II. Responsibilities of faculty members and institutional officials
Principal investigators and research team leaders are responsible for the overall scientific and technical quality of individual projects conducted under their auspices. They must set a positive example by their actions and behavior. They need to be sensitive to social and ethical issues that bear upon their research programs, and they are primarily responsible for meeting all governmental and institutional requirements concerning research in humans as well as the use of laboratory animals, radioisotopes, and other hazardous products, and recombinant DNA. Principal investigators and research team leaders bear particular responsibility for resolving issues relating to health and safety in their laboratories. They are also expected to identify potential conflicts of interest which may bias the research under their direction, and to disclose such conflicts to the University, to funding agencies and to journal editors.
Research mentors and their trainees require frequent, collegial interactions to assure optimal conduct of research. At least one senior faculty member should supervise (with mutual assent) all individuals in a laboratory who are not acknowledged independent investigators. If more than one mentor is supervising and training a junior individual, one of the mentors should be expressly designated as having overall responsibility. Mentors should commit themselves to spend the time required for adequate supervision. Moreover, the ratio of trainees to available mentors should be small enough to encourage close and frequent interactions concerning all aspects of research undertaken by a trainee or junior investigator, including the planning and design, data interpretation and preparation of reports. Trainees have both the right and responsibility to be certain that they are adequately supervised during their research training and that the research itself is performed in a manner which reflects high standards for the responsible conduct of science.
Members of a research team including those who are trainees should meet often enough as a group to review their collaborative endeavors that each member of the team is fully informed of the status of the various components of their work. Such reviews should include the design of experiments, technical considerations, the collection, organization and interpretation of data, and the drafts of research reports. Members of research teams should be sufficiently familiar with the policies and regulations of sponsoring agencies and the University to alert principal investigators when it appears that those policies and regulations are not being met.
Departmental and institutional oversight is also necessary. A senior faculty member, usually the departmental chairman, is responsible for promoting sound research practices at the departmental level. Departments should see to it that the principles governing good research are regularly considered at faculty meetings and are appropriately included into formal and/or informal departmental or sectional activities which involve students and trainees.
The Dean or his designate will conduct an educational program dealing with the responsible conduct of research. If specific concerns or disagreements related to research issues cannot be resolved at the departmental level, the Dean or his designate is responsible for counseling those involved. If allegations of academic fraud arise, the Policies and Procedures for Dealing with Allegations of Academic Fraud at Yale University (Weekly Bulletin and Calendar Sept. 16-23, 1996) will be initiated.
III. Conduct of research
Management of Data. Everyone who engages in scholarly inquiry in the School of Medicine is responsible for collecting and maintaining research data in an orderly and systematic manner which will permit ready retrieval even by those who are not familiar with the intricacies of the research or the habits of the investigator. How this responsibility is most effectively carried out will depend upon the nature of the research and the form which the raw data takes. In the case of quantitative measurements, for example, all raw data including all machine printouts should, whenever practical, be sequentially entered into a bound notebook and maintained in a readily retrievable form. Each entry should be dated and should identify the individual responsible for collecting and entering the data. When investigators decide that raw data must be excluded from further analysis, they should note the specific reasons for the exclusion.
Laboratory notebooks containing raw data should be considered the property of the research team leader who has overall responsibility for the research. Others who have been involved in the work should be permitted to take copies of results with them, but the original, complete set of raw data should be retained in the laboratory where the work was done for as long as possible and - depending on the nature of the data - for at least 3 to 5 years after publication of the relevant results.
Once investigators have published a detailed report of their findings, they should cooperate in making readily available to others who wish to confirm their work any data necessary to replicate the published work, such as complete DNA sequences or minor methodological details, even if a journal had not requested the information or had refused to print it. Although investigators should be as certain as possible about the validity of their findings before making their research public, the School of Medicine does not condone secrecy or uncooperative behavior which unnecessarily delays scientific progress or which deliberately misleads others working in the same field.
Authorship of scientific papers*. The increasingly specialized and technical nature of biomedical research requires that investigators understand and properly fulfill the responsibilities with respect to authorship of scientific publications, especially publications which result from multi-disciplinary collaborative research. Senior faculty members and principal investigators of sponsored research bear particular responsibility for the assignment of authorship to publications emanating from their laboratories as well as for the cohesiveness and validity of these publications.
Authorship of a scientific paper should be limited to those individuals who have contributed in a meaningful way to its intellectual content. Each author must have participated sufficiently in the work to take public responsibility for its content. The first author, although often a junior member of the research team, is usually the person who has performed the central experiments of the project. Often, this individual is also the person who has prepared the first draft of the manuscript. All co-authors should have been directly involved in all three of the following: 1) planning some component of the work which led to the paper or interpreting at least a portion of the results, 2) writing a draft of the article or revising it for intellectual content, and 3) final approval of the version to be published. All authors should review and approve the manuscript before it is submitted for publication.
*These principles are consistent with the uniform requirements for manuscripts submitted to biomedical journals now endorsed by over 400 biomedical journals (New Engl J Med 324:424-8, 1991).
Individuals do not satisfy these criteria for responsible authorship merely because they have made possible the conduct of the research and/or the preparation of the manuscript. Under no circumstance should faculty members add as co-authors highly respected individuals merely as an attempt to increase the likelihood of publication. Thus, heading a laboratory, research program, section, or department where the research takes place does not, by itself, warrant co-authorship of a scientific paper. Nor should "gift" co-authorship be conferred on those whose only contributions have been to provide routine technical services, to refer patients for study, to provide a valuable reagent to assist with data collection and assembly, or to review a completed manuscript for suggestions. Although not qualifying as co-authors, individuals who assist the research effort may warrant appropriate acknowledgement in the completed paper.
Senior faculty members have a special obligation to avoid co-authoring papers which have emanated from work independently generated by their junior colleagues. Senior faculty members should be co-authors only if they have made substantial intellectual contributions to the experimental design, interpretation of findings and manuscript preparation.
Issues surrounding authorship are not only important to those participating in a scholarly project, they are often complex or delicate, and occasionally they are controversial. To avoid disappointment, frustration, or embarrassment, participants should carefully and objectively negotiate and resolve matters relating to authorship as early in the course of a project as is feasible.
IV. Evaluation of scholary research
When asked by journals and research agencies to judge manuscripts and grant applications, faculty members or other members of the University community are expected to provide opinions which are not only proficient but which are also fair and unbiased. Therefore, a faculty member should decline such a request whenever a real conflict of interest arises because the work to be evaluated too closely resembles the faculty member's own current or planned research. Because relatively few individuals are sufficiently expert to review work within a particular field, however, journals or sponsoring agencies are very likely to recruit individuals who may perceive a conflict of interest. If the faculty member believes that a possible conflict exists but that an unbiased review is nevertheless possible, he or she should disclose the nature of the conflict to the journal or funding agency. Potential conflicts must also be disclosed when faculty members are asked to review manuscripts or applications evaluating products of companies in which they have financial interests or when the faculty member has a financial interest in a competing product.
Whether or not the faculty member perceives any potential conflict of interest, all manuscripts or grant applications should be reviewed and returned as promptly as possible and should be treated as confidential. The material should not be copied nor should it ordinarily be shown to others. If during a review it becomes apparent that the opinion of other experts would be helpful, the faculty member responsible for the review should, depending on the policy of the journal or funding agency, either request permission before consulting others or acknowledge the individuals who contributed to the review.
Under no circumstances should faculty members who are asked to review new and confidential information provided in manuscripts or applications use that information to further their own scholarly endeavors before that information has been made public. They have a particular responsibility not to delay their reviews to allow completion of their own related research. Moreover, faculty members who are reviewers should refrain from requesting further information or reagents from investigators who have completed work described in manuscripts or applications before that work has been formally accepted or approved.
This document is current as of November 2007
In addition, refer to extensive information available at the Yale Office of Research
Administration webpage on Responsible Conduct of Research