Should authors have to retract papers based on data obtained unethically?

via Kent State

Careful Retraction Watch readers may have noticed that one of the categories in our right-hand column under “by reason for retraction” is “lack of IRB approval.” That’s because in just over a year, we’ve written a number of posts about two cases of retractions for that reason.

One was the now-infamous case of Joachim Boldt, who has retracted some 90 papers. The other was more mundane, about a group studying injuries among Aussie rules football players

These retractions — and another case in which lung cancer screening trial investigators have said 90 percent of their consent forms are unobtainable, according to The Cancer Letter and The New York Times — raise some important ethical questions that we explore in our latest LabTimes column. Excerpt:

Certainly, violation of research ethics is a profound problem that journals must take seriously. But some observers raised a provocative and interesting point: What if a study lacks IRB approval but produces impressive (and reproducible) results? Is retraction really the only option? Should other researchers be denied the chance to learn from the tainted finding?

Read the column to find out how we answer those questions. And let us know how you would.

12 thoughts on “Should authors have to retract papers based on data obtained unethically?”

  1. There was a Spanish film in the 60’s where a military commander was given a medal for a superb operation in the field and then shot for disobeying orders during that operation (that’s as far as I remember details). Currently, the justice works in weird, illogical ways, mixing everything together under political pressures. No, I don’t believe papers in question should be retracted, unless of course there are other faults material to the results. There should be other methods to punish researchers for violations not material to the veracity of the study. These other sanctions should be acknowledged in the journal.

  2. My (semi-)rhetorical questions that disagree with you (somewhat):

    Since IRB’s were only established in 1974, does why do we not ignore all research published prior to that date? IRB’s are a procedural requirement (and well-justified in my mind), not an ethical one.

    Why are journals (both commercial and non-profit) in the position of policing this requirement? How did they appoint themselves are judges for these retractions, since that position is not legally established?

    As Pyshnov said, we need to find some other method of sanctioning this unacceptable behavior.

    1. “Why are journals (both commercial and non-profit) in the position of policing this requirement? How did they appoint themselves are judges for these retractions, since that position is not legally established?”

      Actually, they can be the judges for these retractions, because that position IS legally established. All journals I know of (there may be some exceptions) require that an IRB is in place upon submission of an article. They have the legal right to put rules for publication in their journal, and if you break that rule, they have the right to remove that publication from their journal.

  3. Marco,

    My point was that the laws do not require journals to police for IRB usage. You stated, and I agree, that journals can do this, but that is something that they have taken on by themselves. If a journal decides to not police this matter, they are also free to do so, as the law does not require them to do so.

    I’m not sure that I am comfortable with journals being the more-or-less sole entity policing the matter, especially since it is in their interest to not have retractions. Are they now going to have authors submit proof of IRB approval? How can that be verified and authenticated? It’s a bottomless quagmire for journals to do this and they have very little incentive for the effort.

    Lastly, you only brought up the second issue I raised, and that is a subset of the first question. IRB approval is not a universally accepted ethical restraint. A tremendous amount of science, indeed the cornerstones on which out understanding of biology and medicine are based, was completed before IRBs. Since journals are taking the self-appoint role of pulling knowledge from the scientific community because there wasn’t IRB approval, why don’t they retract all such publications from before 1974? After all, they have appointed themselves to the role of police officer – they need to be consistent, right?

    As I stated above, I fully support the use of IRB’s, but see them for what they are – a good idea, a procedural and legal requirement, but their use, or lack thereof is not an ethical issue.

    1. It is correct that journals do not need to police these matters, but neither are they required to do a whole lot of other things that they also police. A journal that does not demand IRB approval will rapidly see its approval ratings drop. They may not like retractions, but it is also not in their best interest to be known as the place where people publish non-IRB approved experimental studies. Of course people can cheat, but at least then the journal can wash its hands in innocence, and even start legal proceedings.

      Regarding your first part, yes, a lot was done before IRBs were required. But you can’t retrospectively demand something. That is, to be consistent, they do not need to retract publications from before 1974.

      And in my opinion IRB approval is an ethical issue, since the IRB should check the trial is set up according to proper ethical guidelines.

  4. Now I understand it. Journals should have nothing to do with it. This is 100% responsibility of the hospital and any place that makes patients available to the study. But they tried to spread this responsibility onto all others. This is the work of the activist “groups” (pronounced “grloups”). As in all other such cases, when all institutions are vigilant, in the end, no one is ever responsible.

  5. Sure, the law does not require medical journals to police these things. But surely the research community does want ethical standards upheld? Journal editorial boards, that decide journals’ policy, are representative of the research community. Sometimes they come together, for example the International Committee of Medical Journal Editors – – some years ago defined a set of unform requirements for submissions to medical journals, and these now include ethical considerations.

    SImilarly, the Committe on Publication Ethics – – is a forum for editors to discuss ethical issues. It has produced guidelines on retractions and one criterion for retraction is if a paper “reports unethical research”.

  6. If a child is conceived through rape we have to prosecute and jail the rapist or, in some countries or US states, to execute capital panishment to the rapist, but we should not kill the child who has been already born.

    The aim of the panishment for crime is to prevent other protential criminals from commiting the same crime. To achieve this aim it is enought to prosecute the rapist, but there is no need to prosecute the child.

  7. Back in 2003, the JCI retracted a paper ( that didn’t have IRB approval ( The IRB at the authors’ institution actually was the one to report the authors to the JCI, since the senior author had been on probation (
    I believe the author has been through a lot of training in the interim, but we at the JCI felt very strongly at the time that the paper could not remain published in the JCI based on the reports we had.

  8. The discussion here may confound several quite different questions. None of them has any easy answer, but it may help to distinguish the issues. (1) Are the research results unethically obtained, (2) when, if ever, should unethically obtained results be incorporated into the accepted body of scientific knowledge (3) who decides 1 & 2, and (4) what may they do about it? Bear in mind that “unethical” can and has been used to mean anything: from sadism to inadvertent failure to complete some minor paperwork, or from fraud to political heterodoxy. It depends on the ethical system and the severity with which that system is applied.

    I suspect that the answer to (3) tends to determine the answers to the other questions in most cases. In any event, having the other questions questions addressed by several different groups (e.g. university, IRBs, journals and government) with different ethical slants and overlapping jurisdictions — as currently occurs — may be close to the optimum approach, even if it sometimes leads to inefficient decision-making or inconsistent results. Science and ethics are two different things. They are usually consistent, but it’s precisely in the rare cases where the two aren’t completely compatible that one needs to be careful to think in terms of accommodation — not bright lines — lest one of them attempt to swallow the other.

  9. First anatomists were criminals in their times. I in no way condone “unethical” behaviour when it’s seriously wrong, i. e. hurts people. But in our time, we witness ridiculous, emotionally charged misunderstanding of both law and science.
    “Bear in mind that “unethical” can and has been used to mean anything: from sadism to inadvertent failure to complete some minor paperwork, or from fraud to political heterodoxy.”, as Toby White says. The acceptance of these series of “uniform requirements” by scientific community is the result of unwilling to think and fear. Educating the community, asking questions must continue; it will provoke thought in other areas also. There is no progress in blindly following idiots. And, on the other side, there is no progress in charging sadistic doctors with violating “uniform requirements” when they have to be jailed. What these “ethics”, policies and procedures lead to is smearing and burdening all, but never punishing the guilty.

  10. Honestly, if the authors found it prudent to just disregard some aspects of research ethics (whatever the definition of ethics) why should I trust that the authors followed research ethics on other aspects, like data collection, statistics and presentation? If the results are reproducible, as in independent research by another researcher, why is it a problem to retract this particular one? Even if the statistics is sound using this data, again, why would I trust that the actual data collection was done in the proper manner since the authors seems to show some lack of ethical judgement?

    It may be that the authors personally and others don’t agree with the IRB standards, but that doesn’t really mean that they can do whatever they like.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.