Liz Wager: Do we need to rethink our approaches to research misconduct and research integrity?

Liz WagerYesterday I took part in a joint BMJ/COPE meeting on research misconduct. The discussion set me thinking about factors that create and sustain healthy research environments.

When we talk about misconduct, we often think of the cases that hit the headlines (such as Hwang Woo-suk or Scott Reuben). If we use these examples to suggest that institutions should take research integrity more seriously, they understandably retort that such egregious behaviour is very rare. They also point out that it would be as disproportionate and unhelpful to assume that every researcher is a potential fraudster as to assume that every doctor is a potential mass murderer based on the case of Harold Shipman. So we often reach an impasse and accuse the institutions of complacency and continue muttering to ourselves.

Having failed to start a reasonable conversation about research integrity, based on major scandals, we see no point in trying to address so-called “minor” offences. But as Iain Chalmers said at the meeting, “lesser offences,” such as failing to publish research and publishing welcome results more often than disappointing results, harms many more patients than the high profile scandals (since it distorts the evidence on which guidelines and clinical practice are based). While plagiarism may be a nasty symptom of a sick system, it has probably never killed anybody while unreliable guidelines and misguided research undoubtedly have.

Other speakers addressed the role of institutions in encouraging good practice as well as their role in investigating serious misconduct. Until the meeting, I had always viewed these as one and the same thing. But I’m starting to think it’s more helpful to consider them separately. Viewing research institutions as microcosms of society produces a helpful analogy. Murder, is, like extreme research misconduct, thankfully, rare. But few would argue that we don’t need a police force and judiciary to deal with it when it occurs. However, we don’t expect the police to teach our children good manners and how to be responsible citizens—that’s the role of families and teachers. (Incidentally, most people would also admit that the existence of a penal system can never eradicate crime, and a speaker from the US pointed out that having good investigations doesn’t abolish misconduct.)

To encourage a healthy research climate we need the equivalent of both the police and the parent. Institutions need to have effective and efficient systems for handling allegations of serious misconduct. This isn’t easy and may require outside, expert help or advice. Institutions also need to nurture their employees and provide day-to-day mentoring and training—in other words to be a supportive “family.” If researchers “grow up” in such a positive environment, which inculcates and supports good values, then misconduct, both “major” and “minor” should decrease; but it’s unrealistic to expect that nobody will ever be tempted to break the rules so the research family still needs to pay its taxes to support an effective system for dealing with the problem children.

Thinking of the saying that “it takes a village to raise a child” we shouldn’t forget that a healthy community doesn’t only rely on parents (in this analogy probably research supervisors). Other players such as funders and editors need to be involved … but that might stretch my analogy too far.

Liz Wager PhD is a freelance medical writer, editor, and trainer. She is the current chair of the Committee on Publication Ethics (COPE).

  • Joerg Zwirner

    COPE and in particular the medical journals at large are at the core of the problem.
    Similar to universities they have a conflict of interest and want to preserve their reputation. They should take a more active role, e.g. in spotting image manipulations. The Abnormal Science Blog is full of it. Enjoy.

  • Elizabeth (Liz) Wager

    COPE encourages journals to screen submissions for image manipulation and plagiarism. However, COPE is an advisory body, it is not set up to investigate individual cases of misconduct. Similarly, we do not believe that researchers would be best served if journals attempted to investigate misconduct cases, but we do believe that journals should pass evidence onto institutions (or other organizations where they exist). Most journal editors work part time and are neither trained nor resourced to investigate misconduct allegations except in the most straightforward cases of obvious plagiarism, redundancy or image manipulation; we therefore think the researcher's employer and/or the research funder should take responsibility for this. When problems are found in one paper, it very often transpires that this is not an isolated case, therefore employers should take responsibility for scrutinising previous publications when researchers are found guilty of serious misconduct — another reason why we do not feel this should rest with the journals. However, as you may have seen from my earlier article (BMJ 343:d6586) , we realise there are sometimes problems with institutions investigating cases, which is why we took part in the meeting hoping to strengthen the systems in the UK to ensure that universities are supported and, where necessary made, to do this.

  • Excellent review. Your view about a 'supportive family' is what is lacking in most busy institutions. Mentoring needs a 'whole family' and not just being attached to a supervisor.
    A good sports' coach trains the athletes 'to win but also accept if there is a defeat'. The biggest support in my opinion is teaching and training to accept that a negative result is no problem; it is contribution to knowledge.
    If a researcher accepts rejection of a hypothesis and reports the rejection as such, and we publish it, there will be a great contribution to knowledge as future researcher will be aware of this negative result. There is a general atmosphere (myth?) that editors do not accept 'negative results' or failed hypothesis. The result; all graduate students are presenting only 'accepted hypotheis'. Frankly none of my students or even my colleagues' students has ever reported a rejected hypothesis. Today's student is tomorrow's researcher, therefore the problem is deep rooted.
    Now we need to change this culture. Accept and announce that we print negative studies/rejected hypothesis. A culture of 'truth' in research will automatically develop within two decades.
    Dr. Ahmed Badar
    University of Dammam
    Emame