Last month, a judge recommended that a former University of Kansas Medical Center professor be banned from Federal U.S. funding for two years. The ban came after an investigation showed that the researcher, Rakesh Srivastava, had submitted a grant application that was heavily plagiarized from someone else’s. But there’s far more to the case, as Richard Goldstein –who represented the scientist in Bois v. HHS, the first case to overturn a funding ban by the U.S. Office of Research Integrity (ORI), and who has written about another case for us — argues in this guest post.
Picture this scenario: You submit an NIH grant proposal. Unbeknownst to you, it contains material plagiarized from another scientist. Are you liable for research misconduct?
“The answer is clearly yes.” That’s according to a recent decision by Administrative Law Judge Keith Sickendick, of the U.S. Department of Health and Human Services, in a case involving a former University of Kansas Medical Center professor.
Some of you may recall that in May of this year, Judge Sickendick issued a 126-page decision upholding a debarment issued by the Office of Research Integrity (ORI) against Christian Kreipke. The Kreipke case was significant, I argued in August, because it established that a principal investigator (PI), and corresponding author, can be liable for research misconduct even if the PI was completely unaware of any false, fabricated, or plagiarized material. The decision in the recent case — involving Rakesh Srivastava — appears to increase the likelihood PIs and senior scientists will be liable for research misconduct of which they may be unaware.
As a lawyer who advises scientists on matters of research misconduct, I believe that Judge Sickendick’s decisions in the Srivastava and Kreipke cases impose significant obligations on anyone who puts their name on U.S. Public Health Service (PHS)-supported research. To quote the Srivastava decision, “virtually anyone engaged with an institution applying for or that has received a PHS grant is potentially liable for research misconduct.”
It looks like 2018 is going to be a very important year in research misconduct law.
Here’s What Happened
Srivastava had been a regular NIH reviewer. After an institutional and an ORI investigation, ORI found an NIH grant application submitted by Srivastava contained a “significant” amount of text (between 40 and 50%) from the “Specific Aims” and “Experimental Methods” sections of a grant application submitted by Marcia Haigis of Harvard University. When confronted by these facts, Srivastava denied he knew his grant proposal plagiarized Haigis’s grant application. He claimed post-docs and others assisted him in the application. ORI nevertheless found Srivastava “knowingly” or “intentionally” included plagiarized material in the application and imposed a three-year debarment, a Government-wide ban on participating in any research supported by Federal funding and any other contract with the Federal Government. Srivastava appealed the debarment and the case was assigned to Judge Sickendick.
Although it took almost three years to be decided, the Judge upheld ORI’s misconduct findings. In a 26-page decision, the Judge concluded there were no disputed issues of material fact, thus making a hearing unnecessary. After reviewing the evidence presented by both sides, including a declaration from Srivastava denying misconduct, the Judge concluded there was no question the scientist had knowingly and intentionally committed research misconduct. Judge Sickendick decided a three-year debarment was excessive, and reduced it to two years.
Why Is This Case Important?
Although many aspects of the decision are noteworthy, the big takeaway is that ignorance of another’s malfeasance will not preclude a finding of research misconduct. Srivastava submitted a sworn declaration denying knowledge of the plagiarism. The Judge accepted Srivastava’s assertion, even though there was some evidence Srivastava was the only one who prepared the grant application. However, Srivastava’s professed ignorance of plagiarism was irrelevant. For Judge Sickendick, it made no difference someone else inserted the plagiarised material and Srivastava did not know of, or did not intend to include, plagiarized material in the grant application. All that mattered was Srivastava “intended” to submit to a grant application and the application had plagiarized content.
This is a somewhat expansive reading of the law, and if it is upheld, the decision lowers the threshold for proving research misconduct. If there is plagiarized material in a grant application, publication, poster, or other reported research, an author is subject to a misconduct charge. Ignorance of the plagiarism – even if committed by another – is not an excuse.
Second, although the Srivastava case involved plagiarism, the decision is not necessarily limited to plagiarism cases. Authors whose papers, posters, grant applications, or reported research contain false or fabricated data could also be liable for research misconduct, even if they had no knowledge of any false or fabricated data.
Third, ORI’s litigating position is now stronger because it is easier for ORI to prove research misconduct. The judge in this case decided ORI was not required to prove the scientist knew about or intended to use plagiarized words; all it had to do was prove there was plagiarized material in the grant and the scientist intended to submit the grant. Thus, those accused of misconduct may now choose to settle cases instead of litigating.
Who Is Affected by the Srivastava Decision?
The Srivastava decision has important implications for everyone doing PHS-supported research. Although it is a “plagiarism” case, it has implications beyond using another’s work without attribution. Research misconduct is not limited to plagiarism. It includes using false or fabricated data. Thus, after the Srivastava decision, anyone who puts their name on a grant, poster, article, or publication is at risk for a research misconduct finding if the reported research contains plagiarized, false, or fabricated data. It has especially significant implications for PIs or other senior scientists, i.e., those who rely on others to perform research tasks for them and may not have examined the raw data.
Only time will tell whether Judge Sickendick’s decision will be upheld by a higher court. For now, however, PIs, grant applicants, authors, and every other recipient of PHS funding who puts their names on published work are cautioned to do everything possible to detect and eliminate any plagiarized, false, or fabricated data.
Like Retraction Watch? You can make a tax-deductible contribution to support our growth, follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up for an email every time there’s a new post (look for the “follow” button at the lower right part of your screen), 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].
It is not clear to me how it would be possible for someone to access a grant application to the NIH without being a reviewer on a study section. If Srivastava indeed was not involved in the plagiarism, that still means he shared highly confidential grant applications with others, which already is a big no-no. Plus, it is still his name on the cover page so he should be fully responsible for the entire content of the application, regardless of who prepared it. Things become a little more murky for Multi-PI grant applications.
The remarkable aspect of the decision is the ALJ assumed the scientist was NOT the source of the plagiarized portion of the grant, even though your experience (and mine) suggests the contrary. In effect, the scientist was ‘strictly” (i.e. automatically) liable, event though the was assumed to lack any knowledge of the plagiarism. It seems to me the law protects a scientist from possible debarment and serious damage to their reputation if they have no knowledge or awareness the research or grant application is or could be flawed.
> For now, however, PIs, grant applicants, authors, and every other recipient of PHS funding who puts their names on published work are cautioned to do everything possible to detect and eliminate any plagiarized, false, or fabricated data.
Should this not already be the case always? Ensuring validity and originality is always the responsibility of PIs and corresponding authors. It’s hard to decide where to come down on this — it doesn’t appear to be a misreading by the judge, but it also doesn’t seem fair to hold unaware actors to this level of individual liability.
It has been my previous understanding that the act constituting the misconduct — the generation of falsified or plagiarized material itself — is the act that must be carried out “intentionally, knowingly, or recklessly” to reach the level of research misconduct. In this case, the judge appears to consider submitting the grant itself to have been the deliberate act. It’s unclear to me how the respondent’s actions here with respect to the plagiarized material would be any more than “negligent” if he was indeed unaware of it’s presence. Surely negligent behavior demands punishment and remediation, but this seems a bit severe if that’s all it is. It would have been interesting if the other person who inserted the plagiarized material was conclusively identified — I would hope that person would still bear more responsibility than the PI in this case if he/she exists.
Large, multi-disciplinary collaborations are incredibly common, and I’m just not sure it’s reasonable to expect a PI to be able to personally assess the validity of every single thing that’s being sent to him in such situations. The recent NIH scientist comes to mind whose laboratory was shuttered after multiple separate instances of falsification — those individuals in the laboratory who actually fudged the data were taken in for misconduct, and despite the eventual significant ramifications for the PI, there are apparently no formal findings against him. In fact in my recollection, the images published by the fellow who ended up later working for the cannabis company should have been clearly identifiable as complete garbage by anyone who has ever seen a Western blot before. If that PI isn’t subject to findings, it’s hard to see how this PI should be.
Congress has attempted address this type of thing in financial reporting for publicly traded companies (Sarbanes-Oxley) to help prevent the next Enron. This law has defined “intention” to include situations where the internal control systems which management established to prevent error and fraud were not adequate. The CEO and CFO can be held personally liable for violations. Holding PIs to a similarly high standard might make sense, but if so, the requirements and penalties should be explicit.
There has to be room for some nuances. I suspect that the majority of PIs are haunted by the knowledge that we could all be fooled by a brilliant rogue post-doc (the sort who would lie about what was loaded on the gel instead of doctoring the images after the fact). I also suspect there are a minority of PIs who figure that if they push trainees hard enough and ask few enough questions, they’ll get presented with the data they asked for but not be held responsible for the cheating they implicitly requested. In my mind, the former PI is innocent and the latter guilty.
The particular case here sounds rather egregious – the only way the PI could have not known that 40% of the grant was plagiarized was if he made somebody else do the bulk of the thinking and the writing for him, then put his name on it. That’s another unethical can of worms. Yes it is good training for post-docs to help with grant-writing but they should NOT be used as uncredited ghost-writers. Furthermore, having wasted significant time reading and commenting on terrible grants from well-known PIs that must have been written by trainees, I deeply resent those PIs using the study section to do their training work for them.
All true, but what makes this case even more egregious is that plagiarism was committed on an unpublished and highly confidential document i.e. another grant application. A postdoc doing ghost-writing (or ghost-plagiarizing) would not have had access to the Haigis grant application without help from someone on the study section that reviewed the grant, most likely Srivastava himself.