The Federal Government, through the Public Health Service (PHS) and its Office of Research Integrity (ORI), and the National Science Foundation (NSF), has promulgated regulations defining scientific misconduct and outlining the roles both of research institutions and each Federal agency in assuring research integrity. The PHS regulations (42 CFR Part 50, Subpart A) flow from enacted legislation (Section 493 of the PHS Act), while the NSF regulations (45 CFR Part 689) have no legislative history.
In order to identify the research community's view of opportunities for streamlining, a workgroup of research institution officials was convened. The workgroup was diverse, containing research administrators, research investigators, and faculty members responsible for research integrity programs. Both the legal and scientific professions were represented. This diversity is necessary in order to capture the perspective of both those performing the research and those charged with developing, implementing and overseeing research integrity programs at the institutional level. A listing of the individuals serving on this workgroup is included in Appendix 2.
The specific task assigned to the members was to develop proposals to reduce regulatory burden in the Department of Health and Human Services' (DHHS) program that promotes research integrity and investigates allegations of misconduct in science.
It should be noted at the outset that considerable effort is being undertaken currently by the Office of Science and Technology Policy and the Federal science agencies to develop a uniform definition of the term "scientific misconduct." Since one of the most prevalent sources of regulatory burden is the imposition of different regulatory solutions by different government agencies to the same problem, the issue of differing definitions and differing interpretations of those definitions is considered regulatory burden. However, given the commitment and effort being undertaken at the highest levels of government to develop a common definition that is consistent with the values of the research community, the workgroup devoted its efforts to identifying other aspects of the program that could be streamlined.
The ORI oversees the DHHS program for research integrity and is comprised of the Division of Research Investigations, the Division of Policy and Education, and members of the Office of General Counsel. As the name implies, the Division of Research Investigations assesses allegations of misconduct, monitors and reviews institutional inquiries and investigations at extramural institutions, and conducts investigations in DHHS's intramural programs and occasionally at extramural institutions on an as-needed basis. The Division of Policy and Evaluation develops regulations, policies and procedures, manages the assurance program, conducts institutional compliance reviews, oversees institutional responses to complaints of retaliation against whistleblowers, monitors the implementation of administrative actions, produces publications, and organizes conferences and workshops. Of the full-time equivalents (FTEs) assigned to ORI, 7 are for the Office of General Counsel, 18 are for the Division of Research Investigations, and 9 are for the Division of Policy and Education. Of those assigned to the Division of Policy and Education, about 2 FTEs are devoted to educational activities, although the resources of the other Divisions are also drawn upon for these activities.
The PHS research integrity regulations, as stated in paragraph 50.101, apply to each entity requesting support for a research, research training, or a research-related grant or cooperative agreement under the PHS Act. They require that each entity establish uniform policies and procedures for investigating and reporting instances of alleged or apparent misconduct involving research or research training, applications for support of research or research training, or research-related support activities funded under the PHS Act. More specifically, the regulations require that each applicable institution provide an assurance that it has established an administrative process meeting the requirements of the regulations and that it will comply with that process. These policies and procedures are to be kept current and must be available to the ORI for review. ORI currently reviews the policies of about 200 institutions per year.
The regulations are quite detailed in specifying how an institution should proceed, should there be an allegation of misconduct. In addition, while the regulations state that the institution has the primary responsibility for conducting inquiries and investigations, they go on to say that "…the Department reserves the right to perform its own investigation at any time prior to, during, or following an institution's investigation." Within that context, the regulations require that a research institution's policies and procedures must commit it to a specific list of actions, including that it will:
- Inquire into an allegation of scientific misconduct immediately and complete the inquiry in 60 days unless there are special circumstances, in which case they must be documented. A written report of this inquiry must cover the evidence reviewed, the relevant interviews, and the conclusions.
- Notify the ORI in writing if it decides to initiate an investigation. This notification must occur on, or before, the date that the investigation is to begin and should include the name of the individual against whom the allegation has been made, the general nature of the allegation, and the PHS application or grant number that is involved. However, the ORI should be notified at any point if the inquiry involves an immediate health hazard, the immediate need to protect federal resources, the immediate need to protect the interests of any of those involved, the likelihood of publicity, or if there is an indication of criminal violation.
- Undertake an investigation, where warranted, within 30 days of the completion of the inquiry and keep ORI informed of any developments during the course of the investigation that may affect funding for the individual under investigation, or that is needed to ensure the appropriate use of federal funds or protect the public interest.
- Complete the investigation within 120 days of its initiation. If the investigation requires more than 120 days, the institution must submit a request for an extension to ORI as well as an interim report. If the extension is granted, periodic progress reports are required.
- Submit a final report to ORI upon completion of the investigation and impose appropriate sanctions if misconduct has been determined. This final report must describe the policies and procedures under which the investigation was conducted, how and from whom information was obtained relevant to the investigation, the findings and the basis for the findings, and include the actual text or an accurate summary of the views of any individual found to have engaged in misconduct, as well as a description of any sanctions taken by the institutions.
Upon receipt of the final report, the regulations state that ORI will review the documentation to determine if the investigation has been conducted in a timely manner with sufficient objectivity, thoroughness and competence. It also may request additional information or additional fact-finding on the part of the institution and once it is satisfied, either accepts or does not accept the institution's finding. ORI currently accepts the institution's findings about 90-95% of the time.
As noted above, the ORI also may conduct its own investigation at any time should it determine that such action is warranted. At the present time, ORI is conducting about 5% of the investigations in cases involving possible misconduct at research institutions; in past years, it has been over 20%.
Finally, any individual found to have engaged in scientific misconduct by ORI may appeal that finding to the DHHS Departmental Appeals Board. Evidence beyond that which is currently in the record may be presented to the Board. The Appeals Board action is the final DHHS action on administrative remedies with the exception of debarment. In that case, the Appeals Board decision represents a recommendation to the Deputy Assistant Secretary for Grants and Acquisition Management, who has delegated authority to make final DHHS decisions on debarment.
These regulations are further amplified in policy documents developed by
ORI. The ORI has developed a Model
Policy for Responding to Allegations of Scientific Misconduct. (PDF
- 290 KB) This model policy is a suggested template for use by research institutions
in developing their policies that are reviewed and approved by ORI. The ORI
also has developed Model
Procedures for Responding to Allegations of Scientific Misconduct. These
procedures are intended to assist institutional officials as they conduct
inquiries and investigations and report their findings to ORI.
Identification of Intent of Research Integrity Protections
In order to put potential issues and solutions into the appropriate context, members identified the objectives that must be achieved if a program of research integrity is to be considered effective. It is the workgroup's view that the protections intended by the federal regulations and the objectives of a program of research integrity are to:
- Establish comprehensive programs that educate and sensitize all members of the research community to the issues of research integrity, that assure the integrity of science, and that ultimately reinforce the public trust in science.
- Assure an inquiry and investigation process that balances the rights of both the accused and the accuser with the objective of fairly and swiftly assessing and adjudicating the allegations. This adjudicatory process should draw on the procedural strengths of the legal system without that system dominating what is essentially an academic decision-making process often involving complex scientific issues.
- Assure an inquiry and an investigative process that minimizes the inherent tensions and loss of morale that inevitably result from allegations of scientific misconduct and the subsequent proceedings.
- Strike the appropriate balance between the primary role and responsibility of the research institution to respond to allegations of scientific misconduct and the responsibility of ORI to provide meaningful oversight consistent with the statute. This balance must be such that it allows the research institution to act collaboratively with its scientists to resolve issues.
- Develop a partnership and a trust between the research institutions and the ORI that will allow them to work collaboratively in pursuit of their common goal of promoting research integrity.
The group concluded that a credible policy has to achieve these protections. A loss of credibility will, in the long-term, be detrimental to science and to society's trust of stewardship by research institutions. Any change to reduce regulatory burden should be consistent with the achievement of these objectives.
Identification of Issues of Concern
Following the development of the objectives that a credible research integrity program must achieve, the members identified those issues that were of concern. This list included issues related to the legislation, the regulations and the inconsistency between the agencies on some aspects of the regulations, the interpretation of the regulations, and practices at the research institution that resulted from the interpretation or over-interpretation of the regulations. It should be noted that the purpose of the workgroup was not necessarily to come to consensus. While there was general agreement on the issues and the proposed solutions, there was not necessarily unanimity on each.
- The regulations as currently written effectively require a duplication of effort on the part of the research institution and ORI; in doing so, they compromise the legitimate role of the institution to have primary responsibility for conducting and adjudicating allegations of scientific misconduct. The members of the workgroup concluded that current wording in the regulations, and the current method of reviewing the institution's actions by ORI, effectively provides the institution with the initial, not the primary, responsibility for conducting investigations and adjudicating allegations of scientific misconduct. Paragraph 50.104(a)(6) of the regulations, state that "While the primary responsibility for the conduct of investigations and inquiries lies with the institutions, the Department reserves the right to perform its own investigation at any time prior to, during, or following an institution's investigation." Furthermore, when the institution completes its investigation and makes its finding, ORI reviews the case to determine that it has been conducted "…in a timely manner and with sufficient objectivity, thoroughness, and competence." According to the regulations, ORI may request clarifications and additional facts, and has the unqualified authority to conduct another investigation and/or overturn the findings of the institution. In the members' experience, this re-review forces institutions to emphasize process over substance, lengthens final resolution of the case often to the detriment of the participants, and creates an adversarial relationship between the institution and ORI.
In addition, members also questioned whether reviewing the facts of individual cases (in contrast to the process employed) by ORI staff was the most efficient way for it to carry out its legislative mandate to provide institutional oversight. Other models based on system evaluation and promoting education and best practices might offer a potential alternative and increase overall effectiveness and efficiency.
- Dividing the case rigidly into "assessment", "inquiry", and "investigation" phases limits flexibility and may be inconsistent with the natural progression of the case. There are three identified phases to a case. The first is the assessment phase, which is identified in the ORI model policy but not formally acknowledged in the regulations. Once the official decides that an allegation merits further action, the inquiry stage begins and, as noted above, the regulation and the model policy are prescriptive regarding the purpose of this phase, the time periods, and the content of the written report that is required. The purpose of the inquiry phase is not to reach a judgment on a finding of misconduct but to decide whether an investigation is warranted. If it proceeds to the investigation phase, the regulations and the model procedures are again prescriptive on the requirements for the investigation phase.
The members observed, however, that cases often do not proceed as uniformly as is implied by the regulations. While some cases may proceed in defined stages, the line between "assessment", "inquiry" and "investigation" is not always precise. For example, in some cases, particularly those that are less complex, inquiry and investigation are essentially completed at the same time. Some of the consequences of this inflexibility include:
- The requirement to continue into the investigation stage even if the individual has admitted misconduct in the assessment or inquiry phase.
- Confusion among the members as to whether a new committee should be established for the investigation stage, which would require repeating some of the same activities conducted in the inquiry phase.
- The additional pressure placed on the institution to resist moving to the investigation phase unless there is overwhelming evidence of misconduct because the requirement to notify ORI at that point is seen by the accused as a major break in confidentiality.
The experience of the members is that these inflexible categories lead to inefficiencies and to a lengthening of the proceedings. Flexibility that acknowledges the diversity in types of cases and that would allow a more natural flow would ultimately yield a more efficient process, and allow the institution to better manage the tensions that arise as part of the proceedings.
- Legal mechanisms are available to pursue redress in research misconduct cases which serve as a disincentive to the institutional investigation of allegations of scientific misconduct. Members expressed concern that legal avenues could be pursued that paradoxically may act as disincentives to both institutions and individuals who are participating in good faith in responding to allegations of scientific misconduct. Specifically mentioned were the use of Qui Tam suits under the False Claims Act. In such cases, an individual may file to recover government funds that were misused; if successful, treble damages may be awarded. Extending this to scientific misconduct, an individual might file against the research institution to recover research grant funds that were misused through scientific misconduct, e.g., where the data were falsified. The Department of Justice (DoJ) may join such a suit and use the institution's investigation files to substantiate the act of misconduct in order to recover treble damages from the institution itself. Thus, as the DoJ may use these evidence files, developed by the institution, against it, vigorous pursuit of evidence by the institution is perversely discouraged.
Also of concern is the possibility that participation on inquiry and investigation panels may expose panel members and investigators to personal legal action. Unless immunized and/or indemnified in some way by the institution or the Federal government, and by the State in the case of State institutions, panel members are at potential risk for legal costs in defending themselves in the case of a civil suit, and either settlement costs or verdict costs depending on the outcome. Such legal actions obviously could act as a disadvantage to the ability to recruit faculty members to serve on such panels.
- Insufficient effort is being devoted at both the Federal and institutional levels to conduct formal education programs in research integrity. The members concluded that all segments of the research community would benefit from formal training in issues related to research integrity and the responsible conduct of research. While commending the usefulness of the workshops and other proactive efforts currently being conducted by ORI, a significant potential exists to expand these efforts as well as the efforts being conducted at institutions.
- A better dialogue must be established between the institutions and ORI. Members noted that there is not always a sense of trust between the institutions and ORI. They observed that they often feel uncomfortable asking ORI staff a question about a specific case out of concern that ORI might unnecessarily insert itself in the investigation or that a call in the assessment phase, for example, will trigger the initiation of the official inquiry stage. Much of this arises from the "second-guessing" role described above where ORI must re-review all of the actions taken by the institution and is structured and staffed accordingly. This tension prevents the institutions from drawing on the experience and resources of the ORI, prevents ORI from achieving its goal of working collaboratively with the institution, and ultimately results in a less than optimal process.
Proposed Solutions to Reduce Regulatory Burden
As noted above, in developing solutions, the group emphasized the need to assure, at the outset, that these solutions must not compromise the items identified earlier as the intended levels of protection. In fact, the members are confident that the implementation of these recommendations would provide a stronger and more efficient program. Recommendations from the members are as follows:
- The role of ORI should be fundamentally restructured and primary responsibility for responding to allegations of misconduct should reside unambiguously with the research institutions. The workgroup strongly recommends that the institutions be given the primary role to respond to allegations of misconduct or concerns about research quality and to adjudicate these cases; ORI should conduct independent fact-finding only in rare circumstances, the characteristics of which have been defined and published well in advance of the filing of specific allegations. The members base this recommendation on their experience that the research community is committed to the principle of responding fairly and quickly to allegations of misconduct. In addition, the institution's familiarity with the structure and context of the research programs frequently allows rapid assessment and/or inquiry. The members contend that eliminating this additional layer of Federal review will result in a more timely resolution of the allegation without compromising the fairness of the outcome. In order to effect this change, the following actions need to be taken:
- Amend the current regulations to make it explicitly clear that the research institutions have the primary role for conducting initial assessments, inquiries, and investigations. Currently this statement of primacy is qualified by the assertion that ORI may initiate and/or conduct its own investigation at any time.
- Amend the regulations to make it clear that the review of a case by ORI will normally be a process review only, to assure that the research institution followed its stated procedures and conducted a fair and thorough investigation.
- Amend the regulations to explicitly state the circumstances under which ORI will conduct its own fact-finding. The members understand that there will be exceptional and rare cases where Federal fact-finding may be necessary. However, the nature of the circumstances creating the need should be identified in the regulations. For example, clear and convincing evidence of institutional bad faith or the use of a fundamentally unfair adjudicatory process could and should trigger ORI review.
- Make the Assistant Secretary for Health the deciding official for all administrative actions taken by the Department of Health and Human Services except debarment, which must be taken by the Deputy Assistant Secretary for Grants and Acquisitions Management; in doing so, eliminate the appeal to the Departmental Appeals Board. The members recommend a process similar to that followed by the NSF. The Director, NSF, makes final decisions on administrative action without an independent appeal mechanism such as a Departmental Appeals Board. This seems to operate more efficiently than the current DHHS process without compromising due process.
- Promoting and conducting education and training in the responsible conduct of research and in practices supporting research integrity should be the primary role of the restructured ORI. The members recommend that the resources freed by reducing the unneeded portion of ORI's role in fact-finding be redirected to promote and conduct education and training programs. Included would be:
- Training programs for the entire scientific community regarding issues of responsible conduct of science, including enough examples to clarify what constitutes the level of intended protections, the spectrum of cases, and the level of seriousness of misconduct, as well as good laboratory practices with respect to mentoring, data review, data retention, etc.
- Professional support and assistance for institutional officials using available technology such as CD-ROMs and Web-based programs.
- Identification and dissemination of best practices throughout the community.
- Forums for community-wide discussion and resolution of issues. Particularly useful would be a future dialogue on the advantages and disadvantages of various standards (clear and convincing, preponderance of evidence, beyond reasonable doubt) that could be used to judge the occurrence of scientific misconduct.
Although more can be done, the members acknowledge and appreciate the efforts that ORI has begun in conducting workshops and encourage an expanded program.
The research institutions should be allowed to flexibly manage the process of responding to allegations from assessment through investigation. The current definitions and requirements for each phase should be modified to allow the institution to adapt to the natural flow of the case. Suggestions made by members include:
- Soften the boundaries between assessment, inquiry, and investigation. This would allow activities currently the exclusive purview of the one phase to occur in another if it enabled the institution to better adapt to the progression of the case, and better manage the tensions arising from the case. It would also preclude the establishment of new committees by the institution as the case progresses from the inquiry to investigation phase.
- Postpone the submission of the respondent's name to ORI until the completion of the investigation, except in unusual or egregious situations where the federal government has a need to know.
- Assert that it is permissible to close out a case once an individual has admitted to scientific misconduct and the penalty has been assigned, if the institution has made a reasonable effort to insure that there is no further misconduct.
- Clarify that it is permissible to use the same individuals to serve on both the Inquiry and Investigations Committees at the institution.
- Develop a set of simpler procedures, similar in concept to expedited review available to Human Subjects Institutional Review Boards, for those cases that are less serious or less complicated.
- Expand and clarify the value of the assessment phase and use mediation or alternative dispute resolution, or an ombudsman, as appropriate, when the issue is an interpersonal dispute rather than scientific misconduct.
With regard to the above, the members were mindful of the need to have parameters that put a priority on timely resolution of the case given the disruption of scientific careers. These proposals are not intended to lengthen a case; in fact, the members concluded that it should reduce the time that is necessary for resolution.
An Advisory Committee for ORI should be established and should be comprised of members of the research community; in addition, ORI should broadly communicate its more flexible approach to resolving issues. The members believe that the research integrity program is better served if there is a collaborative relationship between the research community and ORI. An Advisory Committee would provide an opportunity for ORI to gain the research community's views on policy and substantive matters and assist ORI in the continuing development of a healthy dialogue with that community.
The members were impressed in their discussion with ORI leadership with a flexibility that may not be perceived within the larger community. Members suggest that ORI communicate this flexibility widely through newsletters, changes to the model policy and other forms that may be appropriate. The Advisory Committee may be helpful in assisting in communicating this message in addition to these other vehicles. The objective of the members is the establishment of a trust relationship between ORI and the research community that will ultimately yield a more effective and efficient program.
Necessary action should be taken, including legislative action, to assure that the legal system does not unintentionally present a barrier to the investigation of allegations of misconduct:
Conduct a program of continuing evaluation of the research integrity program. The members concluded that the research integrity program, like every Federal program, should be periodically evaluated. While confident that these recommendations will ultimately produce a better system, continual evaluation will yield additional insights. A cost-benefit analysis as part of this evaluation effort would prove useful and should be undertaken. A future evaluation of the appropriate standard of proof might also be considered.
Amend the False Claims Act to eliminate the potential that currently exists for the institution to have its investigation files used against it by those seeking to recover research funds misused through scientific misconduct.
- Provide qualified immunity from tort claims for those who participate in good faith in responding to, or adjudicating, allegations of scientific misconduct.
As stated in the Ryan Commission Report,
the principle should be one of encouraging the pursuit of thorough investigations
of misconduct and protecting individuals and institutions who have participated
in these thorough investigations. Appropriate action needs to be taken to
correct instances in which aspects of the legal system could be used in
a manner that is inconsistent with this principle.
Finally, as mentioned above, the workgroup did not address the issue of the definition of research misconduct, given the significant attention by the Office of Science and Technology Policy and the science agencies. There was concern expressed, however, during the discussion about the practical difficulty of determining the intent of a scientist accused of misconduct and whether there was an "intent to deceive". Members expressed the need for clarification on the findings necessary to support a finding of "intent to deceive" should that be part of the new definition.
Workgroup members are confident that with the implementation of many or all of these changes, the objective of the NIH initiative would be achieved with respect to the regulations governing research integrity – the existing regulatory procedures would be made more efficient without reducing the intended level of protections.