Retraction Watch readers are likely familiar with the U.S. Office of Research Integrity (ORI), the agency that oversees institutional investigations into misconduct in research funded by the NIH, as well as focusing on education programs.
Earlier this month, ORI released data on its case closures dating back to 2006. We’ve charted those data in the graphics below. In 2021, ORI made just 3 findings of misconduct, a drop from 10 — roughly the average over the past 15 years — in 2020. Such cases can take years.
As the first chart makes clear, a similar dip in ORI findings of misconduct occurred in 2016. That was then-director Kathy Partin’s first year in the role, and a time of some turmoil at the agency. In an interview with us then, Partin referred multiple times to the agency being short-staffed. Partin was removed from the post in 2017 and became intramural research integrity officer at the NIH in 2018.
ORI — as has often been the case over the past two decades — is once again without a permanent director. The most recent permanent director, Elisabeth (Lis) Handley, became Principal Deputy Assistant Secretary for Health in the Office of the Assistant Secretary for Health in July 2021.
We asked ORI to explain what’s behind the figures. A spokesperson responded on their behalf.
Data: ORI; graphic by Retraction Watch
What is ORI’s explanation for the fact that there were so few findings in 2021?
After ORI conducts a thorough oversight review for cases in which the institution has submitted its final investigation report, ORI may or may not concur with the institutional findings (that research misconduct occurred or did not occur). ORI also must decide whether to pursue or decline to pursue (DTP) separate administrative actions. ORI may conclude that the misconduct lacks significance (e.g., no published papers or grants) or that PHS funds were not involved and closes the case as a DTP. ORI’s DTP closure of a case is not an exoneration of the respondent, and an institution can implement its own administrative actions based on its determination of the respondent’s research, scientific, or professional misconduct. In CY 2021, ORI closed 41 cases (Finding of Research Misconduct, No Finding of Research Misconduct, and DTP), compared to 48 in CY 2020 (Finding of Research Misconduct, No Finding of Research Misconduct, and DTP).
Data: ORI; graphic by Retraction Watch
Data: ORI; graphic by Retraction Watch
Per the ORI, a declined to pursue (DTP) closure “involves a case in which an institution found that research misconduct occurred and may implement administrative actions against the respondent, but during its oversight review, ORI determined that a separate PHS finding of research misconduct was not warranted.” A “no-misconduct closure involves a case in which the institution conducted an investigation and determined that research misconduct did not occur. Based on a preponderance of the evidence during its oversight review, ORI concurred with the institution’s determination.” And an accession “involves a case that was resolved during the assessment or inquiry stage of the institutional proceeding. Generally, during its assessment or inquiry into the allegation(s), the institution determined that there was insufficient evidence to warrant proceeding to an investigation. Based on its subsequent review of the institution’s assessment or inquiry report, ORI concurred with the institution.” Accessions include cases in which ORI decided the case was outside of its jurisdiction.
What is ORI’s explanation for why accessions nearly doubled, but findings declined by more than half?
As nearly every labor sector has experienced, institutional workflows were altered over the past two years of COVID-related building closures and staffing issues. These closures in part affected the ability of some institutions to complete their proceedings within the time limitations as specified in the federal regulations. Although the number of allegations received over that time has changed little, the delays in receiving completed institutional reports meant that ORI could focus on the full range of potential accession closures. ORI notes that some allegations may involve many published papers or grant applications with multiple respondents from various institutions, further enumerating the complexity of a case and completing its review. The number of accessions for any calendar year does not necessarily reflect the year in which ORI received the allegation, when the institution started or completed its research misconduct proceeding, or when ORI initiated its oversight review. Generally, an accession closure involves an institutional proceeding that did not progress to an investigation, and ORI’s oversight review concurred with the institution’s determination that there was insufficient evidence to warrant proceeding to an investigation. In other cases, ORI may not have jurisdiction (does not involve PHS funded-research or is outside the 42 C.F.R. Part 93 definition of research misconduct). ORI would close such a case while the institution proceeds to an investigation under its own (or other funding agency’s) authority and relevant regulations. ORI closed 52 accessions in CY 2021 after thorough oversight review of the associated allegations and institutional outcomes. The increase in accession closures reflects tireless work carried out at ORI and by institutions.
Is ORI concerned about how long cases typically take to be adjudicated?
ORI recognizes the importance of and is focused on fully addressing allegations of potential research misconduct in the most effective and efficient way possible in accordance with 42 C.F.R. Part 93. ORI also recognizes that ensuring due process and a full, fair, and independent examination of allegations is in the best interest of all involved. It is important to remember that institutional processes take time. Sometimes the investigation must be expanded in scope to consider other possible research misconduct committed by the same or additional respondents or to examine additional papers and grant applications that were not part of the initial allegations. A thorough oversight review, which ORI undertakes when the institutions complete their work, also takes time. ORI hopes to expand its use of technology for file submission, reporting of allegations, and information processing to improve overall processes, efficiencies, and case closure rates in the coming years.
Like Retraction Watch? You can make a one-time tax-deductible contribution by PayPal or by Square, or a monthly tax-deductible donation by Paypal to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].
An example of the type of thing ORI does not consider misconduct, is self-plagiarism, such as the case I reported here… https://psblab.org/?p=611
Essentially, according to ORI it’s perfectly OK to just publish the exact same data multiple times across different journals, thus gaming the metrics system by which all academics are judged.
42 C.F.R. Part 93.103 needs to be rewritten!
Clearly ORI needs more funding and more staff. I am wondering if some sort of a tax on institutions receiving NIH grants could be levied to fund it?
Even cases that have been resolved by the DOJ are still be pending at ORI. One example would be the Sam W. Lee case (misconduct in grant proposals) which was settled with the DOJ but there is still no finding of misconduct from ORI.
Another case that I have been waiting to see resolved is the Xianglin Shi/Zhuo Zhang case at the University of Kentucky which will be interesting reading because to my knowledge ORI have never found two respondents to be responsible for the same research misconduct.
Scotus (re the interesting funding idea): ORI was (is?) actually funded out of a fixed percentage of the annual NIH budget, nice because that’s generally politically immunized. But irrespective of congressional intent, ORI is administratively within – and its budget controlled by – OASH (Office of Assist Secretary of Health) . . . . which is at its own separate mercy of congressional appropriation process. So OASH controls ORI spending, and you can figure out the rest!
Having read the spokesperson’s comments, MEGO (My Eyes Glaze Over).
This RW story is based on the ORI spokesperson’s account, so one wouldn’t think that this had anything to do with the fact that -in its infinite wisdom (and just before the pandemic) – HHS downgraded ORI’s facilities by moving out of its 7th floor space down to the 2nd floor, eliminating and replacing individual offices with open carrels in that process. (Or so I was told then and had asked RW to look into at the time.)
Hard enough to work on confidential files, and d**n near impossible to review physical case evidence which must be secured in the file from. OK, so let’s add Covid to the mix, and it is not hard to see how the geniuses at HHS really screwed over ORI investigations. The facts are easier to understand than the spokesperson’s fine explanation
Fascinating. Thank you for the insight into the weaponization of open-plan offices.
I had heard of the idea of open offices before I left in 2013, but no one then thought it would be implemented by HHS since it was deemed so crazy and unworkable. (BTW: for clarification, my “file from” is ‘iPadese’ for “file room”.). But this had to be a short sighted decision from upstairs at HHS, fully independently from the troops doing investigations.
Given that it was made during the Trump administration, I think there’s a good chance it was done for the sole purpose of frustrating and demoralizing government workers.
Honestly the shocking thing to me about this is the idea that anybody in the government not in senior leadership still had individual offices.
It’s been almost a decade since I left federal service but at the time the cubicles were disappearing in favor of a completely open office, and there was serious talk of hot-desking.
Jake, what do you mean by “hot-desking” and would that in an open environment protect confidential phone conversations or need for spreading across your desk (and complete office) the original physical evidence in rebiewing these cases? Would a “hot desk” protect the scientist-investigator from even a less-than-clever respondent’s lawyer impeaching their credibility in an appeals hearing: “So Dr, how can we be sure my client’s data hasn’t been altered, since he swears those are not accurate”? Finally (and I don’t know the answer to this question), does the NIH recruit professional scientific staff at NIH work in cubicles, and at what GS level?
Hot-desking is where nobody has an assigned desk. First-come, first-serve at the start of every day. It could be combined with open-plan, cubicles, OR even private offices, but let’s face it, it’s most likely to be combined with open plan. During the workday, it’s no less private (could anything get LESS private than open-plan?), but the lack of assigned spaces means that you can’t just lock up your desk drawer or filing cabinet at the end of the day. I have no idea where confidential materials would be stored overnight. Maybe take them home? Yeah, sounds good, I can’t see anything going wrong…
Thanks Adele, I had not heard that term, which describes a come-and-go, pick-up-and-leave, portable work space. I don’t know how they have managed to make things work in the ‘new’ ORI facilities, but at my time (1993-2013) we (all 8-10 of us) worked on multiple case files at one time, confidential records spread, strewn, and stashed all over our individual secure and locked office spaces. I was told the only individual space was the computer forensics room we established. I can fully understand why the HHS spokesperson was unnamed. RW should look into who pushed those decisions downtown.
NSF moved into a new building in 2018. The initial plan was for open office but that was nixed, the argument being that program officers needed privacy when conducting panels, telephone calls with PIs, etc. The support/administrative staff, however, were put into cubicles. And – and I’m not making this up – low grade levels got cubicles with 66” barriers, high grade levels got 80” barriers.
Can somebody explain how come Fazlul Sarkar, one of the most disgraced “scientist” due to the extent of his research misconduct and retractions, cannot be found in the ORI summaries? At the time his papers were flagged, he had at least 4 R01s and also DOD funding. No investigation of Sarkar from ORI? No punishment for the millions of dollars that were wasted in fraudulent research? No accountability? How is this possible? Something doesn’t add up.
http://retractionwatch.com/2020/08/07/cancer-researcher-hit-with-10-year-ban-on-federal-us-funding-for-nearly-100-faked-images/
Wang was initially Sarkar’s PhD student and then he became his postdoc. Why was he targeted by ORI and not the advisor? The PI of the R01 and several DOD grants was Sarkar, and Zhang was not a co-author in many of the retracted papers. I don’t believe Zhang was held responsible by the investigation at Wayne, only Sarkar. It seems that Sarkar got away with the crime and the student paid the price. All very strange.
I spoke with a staffer with the House Investigations and Oversight Committee on Science, Space, and Technology late in 2021 and while he didn’t share any details, my impression was that they might be working on legislation aimed at addressing some of this and on research integrity in general.
In my discussion with him, I shared a list of ideas that might help with research funded by NIH and other US agencies. As I’m not a researcher (which he knew), the list may be misguided, but here it is:
More clear-cut guidance from the ORI about ethics in publishing. Perhaps adopt these as a code of ethics: https://www.nature.com/nature-portfolio/editorial-policies. While this may only be enforced on NIH-funded papers, having this as US policy would nudge US institutions (at least) to follow suit.
Require research to be published in open-access journals; public pays for it, public should get access
Require data retention for 10 years with a custodian (can be institution or for-profit data warehouse) pre-identified in grant requests and published paper
Require data transparency – with anonymization for patient data
Agreement that funded researchers must submit to audit or examination upon request, or grant money is clawed back
Whistleblower protections in writing before grants awarded
Perhaps adopt the “Bik scale” https://scienceintegritydigest.com/2020/01/08/oops-i-did-it-again/
In my experience working on ORI images with Journals, an institutional data retention requirement (by the funders such as the NIH) would be a key first step. But start with the key motivation in ‘science.’ That of course is “retraction” (of occasional discussion by RW watchers), and that solely is the purview of academic/research institutions and (IMHO) not a call for government. But a funding requirement for data retention would make the key step much easier for Journals.
Specifically, Journals can already exercise common sense (and one in their own self interest), by simply requiring the corresponding author to accept, pre-publication, an immediate retraction ‘for cause’, namely if 1) a legitimate question is raised in post publication peer review and 2) the data is then found to be ‘missing.’ No debate about the results, easy-peasey, and all within academic prerogatives. If the journals stepped up to the plate, 70% of ORI’s investigative work (that involving images cases) would evaporate.
Such an institutional data retention requirement by funders would facilitate what publishers can already be doing to protect their investment and inspire coauthors to review the primary data.
The NIH has a new policy on data management that will go into effect in 2023
https://grants.nih.gov/grants/guide/notice-files/NOT-OD-21-013.html
The policy will will require “storing data of sufficient quality to validate and replicate scientific findings” with an expectation that data will be digital or digitized.
Use of a data management system would establish the provenance of data and make it much simpler to identify fraud. Based on the amount of gel and western blot fraud I can only imagine how many bogus graphs and tables of data the fraudsters are churning out.
There are literally hundreds, if not thousands, of fake, fraudulent, or otherwise “misconducted” research published every year. Everybody knows this. It is also clear that many of these cases are extremely obvious, even if others might be harder to identify. The ORI finds three (3) cases in a whole year… That seems like an institution that has no reason to exists if it can’t work more effectively than that.
I suggest you (re?)read the federal regulations defining the limits of ORI’s jurisdiction and authority.
(Indulge my aggressive mock cynicism here: Historically , the ORI was created to sweep up after the annual NIH appropriation funding parade passed each year, much to the then dismay of the academic research community. Are you saying it is ORI’s (and government’s) task to enforce failures to make really the simple steps that academics, research institutions, and journals could painlessly and easily implement to insure their standards keep up with research practices in the digital age?
In addition, ORI lost last year one of its top (for 12 years) scientist-investigators, described by the former ORI Director as having “brought incredible tenacity, a willingness to teach others and to share her knowledge, strategic understanding of the importance of working with others like our Office of General Counsel and RIOs on the successful outcome of cases, and a keen and strong intellect to our work. The outcomes she achieved with her cases are something I wish the public knew about public servants like her. She fought hard to make sure that biomedical research results can be counted upon, by painstakingly chasing down fabrication, falsification and plagiarism in hundreds of cases across multiple disciplines and with thousands and thousands of images.” https://ori.hhs.gov/blog/personnel-announcement-ori-director
Thanks, Anne.