When conduct of a clinical trial at a research site is called into question, the question arises: What happens next? When there have been allegations made that could be characterized as constituting “research misconduct,” certain findings must be made, according to a defined process. Institutional research misconduct policies vary, and some are not sufficiently detailed to describe the steps to be taken. Important issues may be overlooked, such as (1) identifying who must carry burdens of proof associated with the allegations, and what standards of proof should apply, and (2) whether there is an obligation to report the matter to the Sponsor, or the Department of Health and Human Services’ Office of Research Integrity (“ORI”).

Controlling Policies
A first step is to identify which policies will determine the actions to be taken. This requires identifying the funding source of the study. If the study is supported by funds available under the Public Health Service (“PHS”) Act, e.g., funds from the National Institutes of Health (“NIH”), then the regulations found in Title 42 of the Code of Federal Regulations Part 93 (42 CFR Part 93, or “Part 93”) will control the proceedings for those allegations falling within the definition of research misconduct under 42 CFR 93.103. For those allegations not covered by Part 93, either because the study is not supported by PHS funds, or because the allegations do not fall within the Part 93 definition of research misconduct, only the institutional policy will apply. Identifying whether Part 93 controls a research misconduct proceeding has significant practical consequences, among them, whether there is an obligation to notify ORI. 

There are a variety of funding models for clinical trials, thus the funding model may require unpacking in order to determine whether PHS dollars are involved. For example, one complex model might be where a site that is part of a research group receiving funds and/or study drugs or devices from an industry sponsor also receives infrastructure funding from NIH. In such a case, a mixture of PHS funds and industry funds support the study. Another scenario may be one where a central coordinating center receives NIH funding for a program, and the coordinating center engages clinical sites and acts as the financial steward, flowing PHS grant or contract dollars to those sites through its own business office.

Whether an allegation falls under Part 93 or solely under the institutional research misconduct policy is a key determination. Institutions that receive any PHS funding for research, research-training or research-related grants or cooperative agreements, are required to create an institutional research misconduct policy to investigate and report instances of alleged misconduct. The institutional policy must fulfill the requirements of Part 93; however, it is not necessary that these institutional policies be identical to Part 93 in all applicable respects. In the event of differences, they need only to avoid the imposition of conflicting requirements to those of Part 93. For instance, the institution’s definition of research misconduct must include and not conflict with that in Part 93. For purposes of Part 93, “research misconduct means fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results.” Research misconduct does not include honest error or differences of opinion.

Fabrication is making up data or results and recording and reporting them. Falsification differs in that it involves changing or omitting data or results such that the research is not accurately represented in the research record. However, under the policy of the site, other conduct that an institution wishes to strongly discourage also may be defined as research misconduct. An example of such further conduct may be the failure to comply with requirements for the protection of human or animal research subjects. Where PHS funding is involved, the institution must, under 42 CFR 93.300(b), respond to each allegation of research misconduct for which the institution is responsible under Part 93. Therefore, when institutional policy defines misconduct more broadly than Part 93, e.g., where it includes failure to protect human subjects, the question may arise as to whether ORI reporting obligations would be triggered on account of that particular allegation.

Research misconduct proceedings are phased, where certain findings must be made before the matter is eligible to proceed to the next phase. In the first phase, the allegations are assessed for sufficient specificity and credibility so that evidence of research misconduct may be identified for whether they fall within the definition of research misconduct, and for whether Part 93 applies. Part 93 does not specify a time limit for allegation assessment, but does specify that relevant evidence must be secured and inventoried before or at the time of initiating the second phase, the “Inquiry.” The purpose of the Inquiry is to conduct a preliminary review of evidence to determine whether to initiate an “Investigation.”  Only if there has been an Inquiry finding that an allegation warrants an Investigation, does the proceeding move to the third phase. If an Investigation is warranted, ORI, under Part 93, is to be notified; therefore, it should be established before this time whether the allegations fall under Part 93. Once ORI is aware of a research misconduct proceeding, ORI may notify any other agencies that have jurisdiction, such as FDA. ORI may conduct its own research misconduct proceeding, as well. This sets up a conflict between expeditiously concluding the allegation assessment and Inquiry on the basis of preliminary evidence versus conducting these phases in a more thorough and time-consuming fashion, especially if it appears that the allegations may be unfounded or it appears that the allegations, even if true, may not be amenable to proof on the basis of the quality and quantity of the available evidence.

State of Mind, Standard and Burden of Proof, and Honest Error
Assuming that Part 93 controls a research misconduct proceeding, or that an institutional policy modeled on it does so, not only must the conduct under question fit the definition of research misconduct, but it must have been committed intentionally, knowingly, or recklessly.

Additionally, the conduct must represent a significant departure from accepted practices of the relevant research community. Another key question involves the standards by which the required findings are to be made, and who carries the burden of proof. This has a connection to the requirement that the act be intentional, knowing, or reckless in nature — a connection that is often a source of confusion.

The institution has the burden of showing research misconduct. The evidentiary standard for doing so is the preponderance of the evidence (i.e., more likely than not). Because the section of Part 93 that defines research misconduct also includes the statement that research misconduct does not include honest error or difference of opinion, this has been mistaken for a part of the definition, and thus, to be proven absent by the institution by a preponderance of the evidence. Alternatively stated, it has been interpreted, erroneously, to mean that the institution must prove that the conduct was not an honest error. HHS, in the preamble to the Part 93 final rule (see Federal Register, Vol. 70, No. 94, May 17, 2005; p. 28370), clarified that an assertion of honest error or a difference of opinion is only an affirmative defense. This means that it is the person alleged to have committed research misconduct (known as the “respondent”), if shown to have committed an act that meets the definition of misconduct, who must carry the burden of proof to show that it was done as an honest error or resulted from a difference of opinion. The respondent must do so by a preponderance of the evidence.

It is worth noting that honest errors and differences of opinion about the requirements of a protocol and its explanatory operating procedures are, in fact, frequently caused by inadequate management oversight at a research site. Poor management of research projects leads to inadequate training of those personnel responsible for harvesting and submitting data. It also leads to poor communication among the research team, and this contributes to the persistence of erroneous interpretations of requirements. Understaffing for certain functions can lead to corner cutting by personnel who perceive themselves to have been permitted — or even directed — to do so when a need is ignored by management. A rich source of evidence to support an affirmative defense on the grounds of honest error is often found by focusing on the organizational aspects of a study, and on the management, training and resourcing of personnel. Conducting an exploration of these management issues at the Inquiry phase may lead to an institution’s decision to not initiate an Investigation, despite the procedural designation of honest error as an affirmative defense. Under these circumstances, it is vital that the institution assess the impact of its management practices on the project under question, and take steps to discharge any obligations it may have to the research sponsor under the terms of its agreement — which may or may not require disclosure of the existence of preliminary phases of research misconduct proceedings. Both the suspect study and other research projects should be scrutinized, and effective corrective and preventive action should be taken.

It is important for institutional policies to include sufficient detail to drive effective and timely action once an allegation of research misconduct has been made. Part 93 does not place a time constraint on allegation assessment, during which time evidence is to be secured and inventoried, although once an Inquiry is initiated, it is to be completed within 60 days unless the institution can document a reason for exceeding this time limit. A more thorough exploration during the allegation assessment and Inquiry phases of a research misconduct proceeding may negate the need to initiate an Investigation, and may uncover the need for significant improvements in management.