Report on Research Compliance 21, no. 10 (October, 2024)
Sheila Garrity, director of the HHS Office of Research Integrity (ORI), recently spoke to RRC about the agency‘s new rule revising research misconduct regulations, which has a compliance date of Jan. 1, 2026 (see related story, p. 1).[1] Garrity answered all of RRC’s questions—which were not submitted in advance—providing insights into the genesis of the rule, her view of the comments ORI received, the aspect of the rule attorneys who defend accused researchers may appreciate and the tasks that lie ahead for the agency.
RRC: After the notice of proposed rulemaking (NPRM) was released, RRC reported that it generated a lot of negative comments and some requests that you withdraw it and start over.[2] In the final rule, ORI “concurred and revised” or concurred and deleted many provisions or kept language from the old rule. What is your sense of how that process went for you and the institutions? Do you think ORI misread what they needed, or did you reach a good compromise?
S.G.: I was very encouraged that so many people were engaged in the process because that’s what an NPRM is for. Now, ORI had 20 years of experience dealing with cases of research misconduct. So, [ORI] said, ‘Let’s throw the kitchen sink and see what is needed here.’ But ORI is not on the institutional side. So, we took those comments, and we carefully analyzed them all, and then balanced the needs of the institution and the needs of ORI, and I think we came up with a really good compromise.
RRC: Could you talk about what the biggest changes from the proposed to final rule are?
S.G.: One of the biggest changes is the time [to comply]. The effective date is Jan. 1, 2025, but the applicable date is Jan. 1, 2026. And you saw in the comments, people [said], ‘There’s no way we can get ready.’ So, people have a year to get ready, and we’re going to help them get ready. ORI has to get ready. Institutions have to get ready.
RRC: How about changes from the 2005 regulation?
S.G.: One of the biggest changes from 2005 to the 2024 final rule is the appeals process. [Under the 2005 regulation,] an appeal to an ORI finding took a de novo review. So, the administrative law judge (ALJ) didn’t consider that record that was built by the institution and then added to by ORI; it was a de novo review.
Now [under the new rule for] the appeals process, the administrative record starts at the institution. ORI adds to it. And that’s what the ALJ will review. That recognizes the hard work done by the institutions, by ORI. I think that’s a great change.
The definition of research integrity officers was added to the new rule. That recognizes this group of professionals that do this hard work. I think that’s important.
The timelines were adjusted for inquiry time requirements and the investigation time requirements; that allows institutions a little more flexibility. We still have timelines in place. They’re going to have to ask for an extension if they need one.
RRC: Was there anything you wanted to keep in that you did not keep in, for example, requiring transcriptions or the pursuing-all-leads language?
S.G.: No. I was not wedded to those [sections]. You know that I was a research integrity officer and so we really wanted to be careful not to introduce additional burden on the side of the institution or to discourage complainants from coming forward. I agree with the commenters that said these would do just that. So, we changed them.
RRC: Do you think attorneys who defend accused scientists will be happy with the rules? Is there anything in particular you think they wanted that ORI included in the final rule?
S.G.: I do think they will be happy with the rules, [particularly] the appeals process. I think that that’s a big win for everybody.
RRC: There’s a section on assurances and subrecipients. So, subrecipients already have to have an assurance, right? That's not a change, is it?
S.G.: That’s not a change. The change introduced in the NPRM said the prime had to take responsibility for that. And here again, we have to balance the burden issue because institutions can have a grant with hundreds of subs and that would really be onerous for them to track all those assurances. So, we are just reiterating…the subs have to have an assurance with ORI.
RRC: In some instances, the final rule uses “may” instead of shall, and in the case of your option to publish misconduct findings, the NPRM proposed “shall,” but the final rule returns to the 2005 wording of “may.” Does ORI have any intention of discontinuing its practice of publishing findings in the Federal Register?
S.G.: No. But what that [wording] does is give us the flexibility. And I don’t think ORI has ever used that flexibility. [But] say someone died. Would we need to publish a finding [on ORI’s website]? It’s still going to be in the Federal Register.
RRC: Findings that you publish on your website are followed by publication in the Federal Register of the same information. So there’s no change from the 2005 rule.
S.G. Correct.
RRC: You’ve given yourself quite a number of assignments going forward. You’re going to publish guidance, and the rule mentions future policymaking. Is there a difference between those two?
S.G.: Through language used in the preamble, ORI would like stakeholders to know we plan on issuing guidance as part of helping the community prepare for implementation of the final rule. ORI has outlined areas where it intends to issue guidance and may issue additional guidance as needed.
We were very careful in the final rule to say, we “will” issue guidance or “may” issue guidance. We didn’t want to say that for every change we’re going to issue guidance. But some of them need guidance because some of these areas will continue to evolve. Do we want to lock ourselves into putting something in the final rule, or do we say, “Let’s work and be able to issue guidance?”
Policymaking is different than guidance and we are working on responsible conduct of research policies with institutions. We’re working on template policies so that institutions are ready. It's like, “Here’s a template policy that complies with a new rule.” That’s the difference there.
RRC: What are the priority areas for issuing guidance or policymaking? There are areas for which you said you could do guidance—defining the research record, internet definition, what’s a smaller institution, practices for pursuing all leads, considerations for when there are multiple institutions. And then you mentioned policymaking on the assessments and self-plagiarism, which you’re not defining in the new rule.
S.G.: The top priority [is] a template policy that complies with the new rule because institutions have to get ready. And so, this year [leading up to the compliance date] is going to help.
The institutional record…we need to give some guidance [on this]. We changed that from the NPRM to the final rule to say you have to do a general index of everything you sequester [and] have to be specific when you relied on that in making your finding or closing the case.
And why did we do that? Because ORI may need some of the evidence that was sequestered even if the institution didn’t use it. So, we’re walking that delicate line, but not requiring the institution to overly index what they sequester.
1 Theresa Defino, “New ORI Misconduct Rule Offers ‘Thoughtful‘ Revisions in Wake of Highly Criticized NPRM,” Report on Research Compliance 21, no. 10 (October 2024).
2 Theresa Defino, “Penn Calls for Withdrawal of ORI’s Misconduct Proposed Reg; Many Commenters Also Critical,” Report on Research Compliance 21, no. 3 (March 2024), https://bit.ly/3IB2bWC.
[View source.]
