Retraction Watch

Tracking retractions as a window into the scientific process

8 things you might not know about research misconduct proceedings: Guest post

with 20 comments

Callan Stein

Have you ever wondered what could happen if you’re accused of misconduct and face official proceedings? We are pleased to present a guest post from Callan Stein, a lawyer who represents U.S. researchers in misconduct cases, who describes some nuances many may not realize about these situations. 

Most researchers know that being accused of research misconduct is a very serious matter. When research misconduct allegations are made, institutions embark upon lengthy, multi-staged inquiry and investigation processes as required by federal law. The federal government’s Office of Research Integrity (“ORI”) – part of the Department of Health and Human Services (HHS) – oversees those institutional findings and imposes potentially career-threatening punishments on those found guilty. While researchers generally understand the basics of how a research misconduct case unfolds, many are unaware of the nuances that bear greatly on the outcome.  What follows are brief descriptions of eight such nuances of which every researcher should be aware.

  1. While “honest error” exempts researchers from misconduct, it is very hard to prove.

Section 93.103 of the federal regulations explicitly exempts “honest errors” from the definition of research misconduct because “honest errors” are, by their nature, not committed knowingly or recklessly. But while this exemption appears to be a refuge for researchers accused of research misconduct, in practice this is rarely the case because “honest error” is extremely difficult to prove.

As a preliminary matter, the “honest error” exemption is an affirmative defense, meaning the accused researcher bears the burden of proving it. ORI and HHS have also both consistently held that they do not need to disprove “honest error” to make research misconduct findings. Researchers claiming the exemption must do more than simply prove that their errors were unintentional. For example, one HHS administrative law judge required a researcher claiming that a transcription mistake was an “honest error” to prove that the error was not only unintentional, but also that his actions that led to the error were reasonable. Another HHS administrative law judge required a researcher claiming “honest error” to prove that he subjectively believed the error was true at the time of publication.

  1. Accepting responsibility for research misconduct is a mitigating factor, but such admissions will often hurt a court case.

When ORI is imposing punishment for research misconduct, the federal regulations require the punishment to be “commensurate with the seriousness of the misconduct.” To aid ORI in making that determination, Section 93.408 of the regulations suggests seven mitigating/aggravating factors ORI may consider. One such mitigating factor is whether the researcher has admitted his/her misconduct and thereby accepted responsibility for it.

On the basis of these regulations, an accused researcher could reasonably believe that his/her best course is to admit the allegations early. But while doing so may lessen the eventual sentence from the ORI, and earn kudos from colleagues who appreciate someone who takes responsibility for his/her actions, confessing to misconduct is, at best, a double-edged sword when it comes to research misconduct proceedings. First, an admission will not end the institutional process or keep the matter away from ORI. Although federal regulations permit institutions to close research misconduct cases early if the researcher makes a full confession, they also require institutions to notify ORI of the admission in advance of closing the case. Second, institutional regulations often contain shortcuts whereby an accused researcher waives certain rights by admitting misconduct. For example, upon an admission some regulations will permit the institution to bypass the inquiry process altogether and proceed directly to an investigation. This can significantly prejudice the accused researcher in two ways. First, it forces the researcher to defend against a research misconduct investigation that does not have as defined a scope as the inquiry process. Second, because the scope is not well defined, it is far easier for the committee conducting the investigation to expand the investigation to review other aspects of the research. Accordingly, the admission can ultimately result in more significant research misconduct findings and, therefore, harsher sanctions (even after mitigation is taken into account). In fact, admitting research misconduct recently became a lot riskier due to the heightened focus on the potential criminal implications of falsifying research data. This past July, a federal judge sentenced an Iowa-based researcher to 57 months in prison after he pled guilty to federal criminal charges stemming from his falsification and fabrication of research data to obtain NIH grants. Although this is an extreme case, it underscores the seriousness of research misconduct allegations and it is a graphic reminder that researchers should exercise extreme caution when even considering admitting misconduct to an institution.

  1. NIH funding is just one of many things that trigger ORI jurisdiction.

By federal statute, ORI’s research misconduct oversight authority is limited to projects for which federal assistance is provided or requested. The majority of such “federal assistance” comes in the form of grants from the Public Health Service’s granting agency, NIH. However, there are seven other PHS agencies, including the Centers for Disease Control and Prevention, the Food and Drug Administration, and the Agency for Healthcare Research and Quality. Assistance or requests for assistance from any of those agencies will likewise trigger ORI jurisdiction.

Initially, it is up to the institution to determine whether federal funding is involved. But ORI will also make its own jurisdictional determination, and in doing so it has routinely found that a mere request for federal assistance sufficiently triggers its oversight authority, irrespective of whether or not the request resulted in actual federal funding. ORI also does not interpret “funding” to be limited to the payment of salary, but can also include paying for any equipment or other laboratory resources.

  1. You can commit research misconduct by exaggerating or falsifying your own credentials.

Although the federal definition of research misconduct does not specifically address a researcher’s falsification or exaggeration of his/her own credentials when applying for PHS funds, ORI has – in certain circumstances – considered this research misconduct. ORI’s rationale is that reviewing an individual’s credentials during the grant application process is a critical aspect of determining whether that individual is capable of performing the proposed research.

In one case, ORI found that a researcher who held a Ph.D. committed research misconduct by falsely claiming that she was also an M.D. in three PHS grant applications. In another case, ORI found that an individual committed research misconduct by falsely claiming to have attended M.I.T. and earning an M.S. in organic chemistry. Neither researcher was accused of falsifying or fabricating any research data, but in both cases ORI imposed a three-year debarment.

  1. You can commit research misconduct without ever publishing a single word.

When most researchers think of research misconduct, they think of someone who has published false or fabricated results in a scientific journal. But traditional publication is not a prerequisite for research misconduct.

For instance, when a researcher falsifies or fabricates data in a grant application or progress report to PHS or NIH, even if that document is never published, ORI still considers it “reporting research results” within the federal definition of research misconduct. Indeed, the ORI Case Summaries page is littered with research misconduct findings based on false data in grant applications and progress reports. In addition, ORI has likewise made research misconduct findings when false or fabricated data were included in: (1) manuscripts given to a mentor or lab chief; (2) PHS applications for the continuation of already-approved projects; (3) abstracts, posters, oral presentations, and preliminary reports presented at scientific meetings/conferences; (4) central databases (e.g., for multi-center clinical trials); and, (5) patent applications.

  1. Federal regulations make research misconduct proceedings confidential, but only to a point.

Section 93.108 of the federal regulations appears to require that research misconduct proceedings be kept confidential by expressly limiting the disclosure of the identities of respondents and complainants to “those who need to know.” Section 93.300 of the regulations then appears to heighten that requirement by stating that PHS expressly holds institutions responsible for ensuring that confidentiality.

However, almost as casually as PHS bestows this confidentiality, it begins to chip away at it with exceptions that all but swallow the rule. First, institutions must still disclose respondents’ and complainants’ identities to ORI. Second, anyone contesting a research misconduct finding by ORI will necessarily have to forgo this confidentiality because HHS administrative proceedings must be public. Third, and most importantly, the federal regulations’ confidentiality provisions are expressly limited to “research misconduct proceedings.” Therefore, if by committing research misconduct a researcher also violated an institutional policy or procedure (which is likely), there is nothing constraining the institution from disclosing confidential information related to its own internal corrective action.

  1. Researchers should not wait for a hearing before HHS to defend themselves, because HHS hearings are virtually never granted.

Research misconduct proceedings are often lengthy, and can last for months — or even years. For this reason, accused researchers are often understandably hesitant to invest the resources necessary to defend themselves at the institutional stages, and instead rely on their perceived right to a hearing before HHS to appeal a finding of research misconduct. This is a critical mistake. Although Subpart E of the federal regulations purports to provide researchers the “opportunity to contest findings of research misconduct and administrative actions,” in practice this “opportunity” is almost entirely illusory.

Prior to 2005, a panel of scientists adjudicated research misconduct cases before ORI, including granting and denying hearing requests. But in 2005, PHS amended its regulations to impart those decisions to an HHS administrative law judge; since then not a single judge has granted a hearing request in a research misconduct case. (One researcher obtained a hearing but only after he filed a federal lawsuit against HHS to overturn the judge’s initial denial of his hearing request.) See Bois v. U.S. Department of Health and Human Services, et al.

  1. A debarment does far more than just prevent you from applying for federal grants.

Section 93.205 of the federal regulations defines a debarment as a “Government wide exclusion….from eligibility for Federal grants.” Debarred individuals are forbidden from participating in any “covered transaction” with the Federal government, which is defined very broadly — participating in both procurement and nonprocurement transactions such as grants, cooperative agreements, contacts, subcontracts, scholarships, fellowships, loans, and any other form of Federal funding.

From the face of these regulations it is clear that debarred individuals cannot apply for or receive any salary or other financial assistance from the federal government. But because the regulations are written so broadly the federal government has interpreted them to also prohibit debarred individuals from receiving nonfinancial assistance. This means that, in addition to being forbidden from applying for their own grants, debarred individuals also cannot use equipment, laboratory space, laboratory personnel, or any other resources that the federal government paid for, in whole or in part. Put another way, according to the federal government’s interpretation, if you are debarred and your laboratory uses a microscope that was even partially paid for with federal funds, then you cannot conduct research there.

In conclusion, readers should note that these descriptions do not comprehensively address each topic. Researchers accused of misconduct are strongly encouraged to retain legal counsel early in the process, and devise a specifically-tailored defense strategy that incorporates these and other considerations.

Callan Stein is a litigation associate at Donoghue Barrett & Singal, P.C. in Boston, Massachusetts. He has represented researchers all over the United States in research misconduct proceedings at the institutional and ORI levels. In addition to research misconduct cases, Stein’s practice primarily focuses on white-collar criminal defense, and civil and whistleblower litigation.

Like Retraction Watch? Consider supporting our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, and sign up on our homepage for an email every time there’s a new post. Click here to review our Comments Policy.

Written by Alison McCook

August 13th, 2015 at 11:30 am

Comments
  • Miguel Roig August 13, 2015 at 11:58 am

    Elements of numbers 2 and 8 were somewhat surprising to me. With regard to number 8, some (many?) bio-medical departments are partially or almost entirely funded by indirect costs accrued from federal grants. Therefore, it sounds as if debarred offenders would not be able to do ANY kind of work in those departments, including teaching undergraduates, because secretaries, books, and even stationary may have been funded with federal support!

  • Kristen Grace M.D., Ph.D. August 13, 2015 at 12:05 pm

    This comment relates to the statement “Second, institutional regulations often contain shortcuts whereby an accused researcher waives certain rights by admitting misconduct. For example, upon an admission some regulations will permit the institution to bypass the inquiry process altogether and proceed directly to an investigation. This can significantly prejudice the accused researcher in two ways. First, it forces the researcher to defend against a research misconduct investigation that does not have as defined a scope as the inquiry process. Second, because the scope is not well defined, it is far easier for the committee conducting the investigation to expand the investigation to review other aspects of the research. Accordingly, the admission can ultimately result in more significant research misconduct findings and, therefore, harsher sanctions”. The regulations that PHS funded institutions must follow when investigating allegations of research misconduct involving PHS funded research state the following:

    Sec. 93.310 Institutional investigation.
    h) Pursue leads. Pursue diligently all significant issues and leads discovered that are determined relevant to the investigation, including any evidence of additional instances of possible research misconduct, and continue the investigation to completion.

    The inquiry is a preliminary fact finding process which is meant to establish that the allegation is sufficiently credible and specific so that potential evidence of research misconduct may be identified and that the allegation falls with in the definition of research misconduct as defined by the regulation 42 CFR 93 and falls under ORI jurisdiction (PHS funded). It is not a process that is meant to define the scope of the investigation. ORI encourages, and indeed expects investigation committees to broaden the scope of the investigation as necessary throughout the investigation. Making an admission earlier in the process has no bearing on the scope of work the investigation committee may review, but it may, and often does, allow the investigation to terminate early.

    Kristen Grace MD, PhD

    • Callan Stein August 13, 2015 at 12:27 pm

      Thank you for your input Dr. Grace. In my experience I have come across institutional rules that permit the conclusion of an inquiry without a written inquiry report (which would otherwise have been required) when the researcher in question admits to committing misconduct. Although I agree that the Investigation Committee is certainly not bound by the findings in an Inquiry Report or otherwise relieved of any of its duties by it, as an attorney representing accused researchers I find the Inquiry Report to be an invaluable document in effectively and efficiently defending researchers at the Investigation stage and beyond. In addition, I have found the opportunity to present comments on a Draft Inquiry Report to be equally invaluable, especially since those comments are frequently the researcher’s first opportunity to present his/her side to ORI. It is possible that researchers will place themselves in a more difficult position if an institutional policy exists that deprives him/her of those opportunities if he/she makes an admission. There may be times when it makes sense for a researcher to admit to misconduct (and this is certainly not meant to be a blanket admonition on doing so), but he/she should be fully aware of the admission’s ramifications which may vary greatly depending on institutional rules.

      • Kristen Grace M.D., Ph.D. August 13, 2015 at 5:58 pm

        PHS funded institutions must have policies and procedures that adhere to 42 CFR 93. An institution must provide ORI with an inquiry report if it is to continue to an investigation. When a respondent admits to falsifying PHS funded research that is sufficient grounds to for an institution to decide to move to an investigation. They still must provide a report to ORI.

        • Callan Stein August 14, 2015 at 8:19 am

          Dr. Grace,

          For example, I have had at least one case where instead of drafting an Inquiry Report (which would include the Inquiry Committee’s preliminary findings of fact, among many other things required by the Federal Regulations) the institution simply used the researcher’s alleged admission (which was in writing) as the Inquiry Report. The Committee did not draft anything and the alleged admission, which was drafted by the Researcher, did not contain the required Inquiry Report elements. This created a difficult situation, and was a significant battle with the institution, one we ultimately lost when it neither changed its position that no Inquiry Report was necessary nor did it agree to consider the possibility that its process was not in compliance with the Federal Regulations. This is the type of situation to which I am referring.

          • Alan R. Price October 24, 2015 at 8:01 pm

            Note that the 2007 ORI Sample Policy, written by counsel after the updated 2005 HHS (ORI) regulations were implemented, stated:
            “However, if a legally sufficient admission of research misconduct is made by the respondent, misconduct may be determined at the inquiry stage if all relevant issues are resolved. In that case, the institution shall promptly consult with ORI to determine the next steps that should be taken.” http://ori.hhs.gov/sites/default/files/SamplePolicyandProcedures-5-07.pdf (Section V.F)

    • Kristen Grace October 25, 2015 at 9:34 pm

      Thank you for that addition Alan.

  • Christopher Haggarty-Weir August 13, 2015 at 12:18 pm

    Great read!

  • Ken Pimple August 13, 2015 at 2:23 pm

    My thanks to Mr. Stein for his insights and advice. However, I was taken aback by two of his statements.

    In points 4 and 5, Mr. Stein implies that researchers are likely to be surprised that “a researcher’s falsification or exaggeration of his/her own credentials” can be research misconduct (4) and “traditional publication is not a prerequisite for research misconduct” (5).

    In the Federal Research Misconduct Policy of 2000 (https://ori.hhs.gov/federal-research-misconduct-policy), the very first sentence in I. Research Misconduct Defined reads:

    “Research misconduct is defined as fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results.”

    Only the last clause alludes to “traditional publication.” “Proposing” clearly covers grant applications, “performing” presumably includes taking notes or annotating data, and “reviewing” surely applies to reviewing grant proposals as well as peer review of journal and conference proposals.

    Some aspects of the policy may be difficult to decipher, but this does not seem to be one of them. Should not every researcher, who could possibly be accused of research misconduct under this policy’s purview, know and understand the heart of the policy?

    • crh August 14, 2015 at 9:37 am

      Honestly, I would have thought, “you can get in trouble for lying in a grant application,” would be obvious to everyone, without so much as glancing at the official policy.

  • scotus August 13, 2015 at 2:33 pm

    Very well written and informative but a cynic might view this as a thinly disguised advertisement for the legal profession..”Researchers accused of misconduct are strongly encouraged to retain legal counsel early in the process, and devise a specifically-tailored defense strategy that incorporates these and other considerations.”

    This will not be inexpensive. I wonder if it would be possible to get research misconduct insurance to cover the cost of all this legal assistance?

    • Callan Stein August 13, 2015 at 3:24 pm

      The cost of an accused researcher hiring legal representation is a huge issue, and one that really puts the accused in a financial bind. I would strongly encourage anyone who is accused of research misconduct to at least contact legal counsel for a consultation (which can usually be obtained free of charge). Even if a researcher cannot afford to ultimately hire the attorney, he/she should get valuable information from that type of consultation. And depending on what the attorney says during that consultation, the researcher may decide it is a sufficiently serious issue to warrant some form of financial commitment to getting proper representation. Most attorneys who handle these types of cases should be willing to work with researchers within their budget, and researchers should certainly feel free to inquire about any cost-controlling options that may be available (for example, having the attorney utilize lower-cost associates for certain tasks). But with that said, the commenter is absolutely correct that properly representing a researcher accused of misconduct can be expensive.

      Whether or not it sounds like an advertisement, defending research misconduct allegations can be expensive because such allegations are extremely serious. They almost always present the possibility of a federal debarment which can all but end a career. Given these stakes, accused researchers should at least contact an attorney so they can decide for themselves whether the risk is great enough for them to take on the financial burden of legal representation.

      Lastly, it has been my experience on numerous occasions that accused researchers who are reluctant to hire legal representation at the beginning of the proceedings (instead, believing they can address and resolve the allegations on their own) can end up making the situation worse while representing themselves and then end up contacting an attorney later in the case (often after an Inquiry or even Investigation Report is issued finding evidence of research misconduct). By this point the institution has already at least begun to make its decision and it can be very difficult to get it to change its mind. Accordingly, the researcher ends up hiring an attorney at a stage when he/she is unable to have the greatest possible effect.

  • LadyProf August 13, 2015 at 6:09 pm

    Well, but the thing about honest errors is that honest people will accept and admit their error and do the right thing. They will also go to great lengths to make sure said error is never repeated. Sure, one can never prove an honest error, but one’s behavior upon coming to know about the error will have a great bearing on whether it is ultimately determined to be an honest mistake.

    • Callan Stein August 14, 2015 at 8:24 am

      I would tend to agree that individuals who do commit honest errors may typically be more likely to come forward to admit their mistake and accept responsibility for it. However, it is absolutely critical for researchers to understand that honest error and accepting responsibility for a mistake are not necessarily two sides of the same coin. Researchers should not be under the misconception that an admission and acceptance of responsibility will be accepted as evidence (either by the institution or ORI) that an error was, in fact, “honest.” That is an entirely different determination, and one that is often made without regard for the researcher’s acceptance/admission. Indeed, accepting responsibility is identified by the Federal Regulations as a “mitigating factor” for ORI to consider when issuing punishments for research misconduct.

  • Dave August 13, 2015 at 9:39 pm

    Very interesting indeed.

  • Paul Brookes August 14, 2015 at 9:31 am

    A great read, and kudos to RW for posting this viewpoint from “the other side of the fence” (since so much of what usually appears here is from the accuser’s viewpoint, not the defendant).

    I think it’s important to point out, however, that this post only covers a very narrow sub-section of what to do when accused – when the accusations come via an institution or the federal ORI. The number of misconduct allegations that actually make it this far up the scale, is very small. ORI makes official findings in maybe a dozen cases a year, and while there are no figures on their “intake rate” I would guess it’s in the tens or the very low hundreds. So, the actual audience for this post is not very large.

    There’s another, much larger, class of accusations that many readers of this site (and PubPeer) will be intimately familiar with – i.e. online reports of apparent problems in published papers (aka post-pub-peer-review). On that side of the story, lawyering-up is generally seen as a negative, and often involves SLAPP type threats, where the objective is to make the other party shut up rather than actually prove the accusations are ill-founded. The result is often a Streisand effect. A popular rebuttal to this type of threat is “real scientists let the data speak for themselves, and don’t need lawyers to do it for them”.

    Perhaps RW would entertain an op-ed from another lawyer, of the type who espouse defamation suits and similar threats against websites and their commenters. It would be really interesting to see how such a person would justify lawyering-up as an appropriate response to objective public discussion of scientific data. Obviously defamation law and white collar criminal defense are very different legal arenas, so a guest post from a different type of lawyer would probably be informative.

    • Jerryjoe August 14, 2015 at 3:54 pm

      You’re way, way off about ORI’s “intake,” Paul. They review everything (at least everything that moves to an ‘official’ Investigation, to my knowledge); even if they don’t end up making formal findings for one reason or another. Often they just decide to sign off on the findings and action taken by the institution. Their workload is actually quite large though; I’m surprised you have this misconception. Surely you’ve talked to them often?

      Second, much of the rest of your comment sounds dangerously like “if you’re innocent, you have nothing to hide,” which as I’m sure you know is just a terrible, juvenile argument. Don’t let me put words in your mouth, but let me remind you of the words of Cardinal de Richelieu, “If one would give me six lines written by the hand of the most honest man, I would find something in them to have him hanged.” I agree with the guest’s implication that a wise respondent (we don’t call them ‘defendants,’ surely you’re familiar with the lingo as well?) will enlist the help of an attorney early-on. There’s very good reasons anyone accused of a crime is guaranteed legal representation in the US, and while these things rarely rise to the level of criminal proceedings, one’s livelihood is very much at risk and in an imperfect world, the wrongly-accused can’t simply rely on the integrity of the proceedings and the honesty of the other players.

  • Miguel Roig August 14, 2015 at 11:39 am

    As the readership of RW is multi-national, I wonder if the title of the post can be amended to reflect the fact that processes described are unique to the US. In that respect, and in reaction to Paul Brooks’ suggestion for another point of view, it would also be very instructive to have an analogous post from knowledgeable individuals from other nations, particularly those that contribute significantly to science (e.g., China, India, or Brazil). What happens when a scientist from one of these nations is accused of committing research misconduct?

  • Robert Geller August 14, 2015 at 7:20 pm

    ORI regulations mention specifically that honest errors or differences of opinion are not misconduct. The original post discussed honest errors, but not differences of opinion. Does the author have anything to say about differences of opinion?

    • Callan Stein August 17, 2015 at 9:55 am

      According to ORI guidance, differences of opinion are treated identically to honest errors in that it is an affirmative defense that must be proven by the accused researcher. Other than this, there is scant guidance on what constitutes a difference of opinion that is exempted from the definition of research misconduct. Based on ORI’s guidance concerning honest error, however, it seems likely that an accused researcher would have to establish, at a minimum, that: (1) his/her interpretation of the results that led to the difference of opinion was both genuine (i.e., he/she actually interpreted the results in that manner, and did not do so solely to dispute research misconduct allegations), and (2) his/her interpretation is objectively and scientifically reasonable. Put in a more practical sense, it is likely that the difference of opinion exemption has been included to prevent competing researchers from accusing each other of misconduct simply because they disagree about the interpretation of certain research results.

  • Post a comment

    Threaded commenting powered by interconnect/it code.