Skip to Content

The Scientific Literature

Competing Interests

The Nature publishing group recently changed its rules on author disclosure for things that might affect objectivity in the authorship of papers. The criteria include the usual ones regarding financial interests, employment, funding, and so on. There’s a new section on non-financial stuff, though:

Non-financial competing interests can take different forms, including personal or professional relations with organizations and individuals. We would encourage authors and referees to declare any unpaid roles or relationships that might have a bearing on the publication process. Examples of non-financial competing interests include (but are not limited to):

  • Unpaid membership in a government or non-governmental organization

  • Unpaid membership in an advocacy or lobbying organization

  • Unpaid advisory position in a commercial organization

  • Writing or consulting for an educational company

  • Acting as an expert witness

I see where they’re coming from on these, but this does wander off into a gray area. I think, in theory, that authors should indeed disclose advocacy or advisory positions that could be seen to affect their objectivity, but there’s a chicken-and-egg problem: what if a scientist comes to a viewpoint based on their research, and acts on those conclusions by joining an advocacy group? Has membership in such a group affected their objectivity, or is the wrong way around?

Note how this is worded, as well. “We would encourage”, but up at the top of the page, it says this:

In the interests of transparency and to help readers form their own judgements of potential bias, Nature Research journals’ require authors to declare any competing financial and/or non-financial interests in relation to the work described. The corresponding author is responsible for submitting a competing interests’ statement on behalf of all authors of the paper.

So that makes it sound as if the non-financial interests are also required, but when you get down to the language of that section, it’s not so clear. There’s an editorial (open access) at Biocentury that wonders about this even more:

Under these criteria, the author of a paper about climate science would need to disclose his leadership position in the Sierra Club. A physician publishing research on the safety of a new contraceptive might be expected to reveal that she had advised Planned Parenthood.

The broad criteria for disclosure in Nature’s policy could easily be interpreted to include religious beliefs; some journals explicitly demand that authors provide information about their religious convictions.

I’d be interested in knowing which journals those are; I’ve never encountered anything like that, I have to say. But although this might seem to be taking things a bit far, the sorts of requirements that Nature (and others) are adding could be abused in this way. The lines between political, religious, and other beliefs are (and have always been) blurry, and trying to unblur them can lead to real trouble. There were plenty of ugly examples during the 1950s and 1960s, for example (as the editorial points out) what Linus Pauling went through for his nuclear disarmament views.

Pauling was not a Communist nor a Soviet agent, but he got tarred with that brush. The whole ideological history of that era is a mess, of course. There were people like Pauling, but there were also people who were accused of being Soviet agents who really were, as things turned out, and you can still start fights over who belongs on which list with people of certain ages or political affiliations. I worry that expanding authorship disclosure in this direction could lead to more rounds of orthodoxy-sniffing and purity tests if we don’t watch out. Editorials are one thing – but a scientific paper’s results should stand up to scrutiny (and stand up to reproduction) on their own, regardless of the beliefs of its authors.

23 comments on “Competing Interests”

  1. Steve Usdin says:

    Here’s a link to the World Association of Medical Editors statement on COI: http://www.wame.org/about/wame-editorial-on-coi. It includes: ” (d) Political or religious beliefs
    (e) Institutional affiliations.”

  2. Chad Irby says:

    One of my favorite disclosure slides:

    “I’m not important enough for anyone to try to influence my research.”

  3. Steve Usdin says:

    Here’s another example. The box listing examples of interests in biomedical research starts with: “Personal, religious, or political beliefs.” http://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.2001221

  4. Adam says:

    One could argue that publishing in Nature (or Science, Cell) now contains an inherent conflict of interest: if one is able to publish in Nature, one will make career gains in various forms (grants, professorships, promotions, press etc.) not usually associated with publishing in “lower” journals.

    1. RM says:

      I’m continually amused by the [pharmaceutical company] researchers with [pharmaceutical company] affiliations publishing research on [pharmaceutical company]’s compounds, which more often than not are “The authors declare no competing financial interest.”

      It’s been explained to me that normal employment arrangements typically don’t fall into the “reportable conflicts” category. Journals generally take the stand that the affiliation line suffices for employment-related disclosures, and it’s caveat lector regarding standard employment-related bias/conflict of interest. The competing interest section is only for things that a typically reader would otherwise be unaware of. From that sense, I can kinda see the point of the new disclosure requirements. The publish-or-perish conflicts are implicit with an academic affiliation, but the fact that they spend time promoting [cause X] may not be readily apparent.

      1. Isidore says:

        The same sort of thing applies to working at a university, does it not? A publication in a prestigious journal would likely have an impact on an assistant professor’s chances of getting tenure. Not to mention one’s chances of getting one’s grants renewed, which in many cases pay for part of one’s salary.

  5. Chrispy says:

    Adam has it right. Frankly, the upside of publishing fiction in these journals has long exceeded the downside of getting the science wrong. This effort to increase disclosure is insufficient to address the fundamental irreproducibility of MOST of what these journals publish.

    These disclosures will be more interesting for what people fail to claim than what they claim.

  6. Sebastian says:

    When evaluating non-financial COI, it takes a smart reader to weigh the evidence with any potential bias from the author. Good readers can usually pick out when data matters even with conflict & when conflict is skewing the author’s viewpoint.

    I’d like to think that Nature has smart readers, but experience has taught me otherwise.

  7. RM says:

    I’m a little mixed on the chicken-and-egg issue Derek points out. I think you could spin that with financial interests too. What if a scientist comes to a viewpoint based on their research (e.g. the benefits of a particular treatment approach), and acts on those conclusions by founding a company to deliver those results (because no one else would)? Ideological goals can drive bias as well (or better) than financial goals can.

    I guess I’m coming from this from my experience on the internet, where a frequent argument is “they’re funded by/associated with big pharma!”, which is treated as a damning put-down. Simultaneously, though, the same people are quoting and referencing Dr. Oz, various naturopathic magazines, self-help books, etc., completely ignoring the vested financial interest those sources have in pedaling woo and making their readers nebulously fearful of “allopathy”. There’s some strange mental block which makes people think “big pharma” is just in it for the money – which makes them and everything they touch evil – whereas the naturopathic industry is in solely for enlightened benevolence – which makes them pure and incorruptible. The reality, of course, is mixed. “Big pharma” is indeed out to make money and does sometimes make shitty decisions because of it, but there is a genuine desire by their employees to help people. Some naturopaths may indeed have a benevolence behind their actions, but there’s plenty out there who are just trying to make a buck, and even the well-intentioned may be suffering from confirmation bias in their zeal, deliberately ignoring countering evidence.

    I think the solution is not to view anything but “no competing interests” as a castigation, but simply additional info about the potential biases of the authors – over and above their affiliation lines. Note that’s *potential* biases. Just because you’re affiliated with the Sierra Club doesn’t mean you’re actually biased on climate change, just like being part of “big pharma” doesn’t automatically make you a greedy, poison-pushing megalomaniac. I don’t think it helps anyone to one-dimensionalize these things.

    1. Old Pump kicker says:

      Unfortunately, we live in a time when many politicians and news commentators make their entire salary by one-dimensionalizing things.

      1. neo says:

        You just broke my irony meter.

  8. Uncle Al says:

    Universal statement of compliance: “The author is an aßhole.”
    How does that leave anybody unsatisfied?

  9. Wavefunction says:

    I see that this could be a problem, especially since people are more likely to be part of advocacy organizations in these politically turbulent times. For instance, should I disclose my activities in the March for Science group? Would simply protesting in their march once or twice be enough to be categorized as “advocacy”? Or if I am attending a local political meetup regularly discussing science policy, should I disclose this fact? I can perfectly see how someone might not want to announce such affiliations in a public journal.

  10. z says:

    Terrible idea. We live in a time where half the population will automatically dismiss whole-cloth everything about a person because they are on the “wrong” side of whatever the issue is (and the other half does the same from the other direction).

    Scientific ideas should stand or fall on the data. Obviously people are influenced by their opinions, but if their work is rigorous enough to merit publication, let it out with no strings attached. If it turns out to be wrong, people will be glad to tear it apart on its merits. We shouldn’t encourage people to lazily resort to their biases to tear things up.

  11. Lawrence says:

    But, as we all know, iirc, a number of times the data is rigged just to get the result(s) desired by the researcher(s). Some (many) have been noted in these very electronic pages!!!!

  12. Tuck says:

    It seems a wise idea. In the nutrition sphere, there appears to be a wide-spread CoI with members of the Seventh Day Adventist Church. This has been detailed recently by Gary Fetke, an Australian doctor who wound up on the wrong side of an argument there.

    The entire profession of dietician was apparently started by an Adventist associate of Kellogg the cereal magnate (himself an Adventist).

    The undeclared CoI is apparently widespread to this day, as the Adventists view this as a form of evangelism.

    I don’t have a problem with that sort of evangelism, but it should certainly be disclosed.

  13. Mad Chemist says:

    So now we’re going to have to disclose our ACS membership if we publish in Nature because it could influence our understanding of the chemical reactions in our paper and our interpretation of the reaction mechanism. XD

  14. Mark Thorson says:

    I want to know whether you’re a dog-person or a cat-person. Dog people are always so prejudiced. Also, whether you’re left-handed or right-handed. Left-handed people are usually gay.

    1. Lawrence says:

      As are people who are happy!!

  15. Jimbop says:

    No ones cares about due credit anymore…the next nobel prize winner was a theif and you will never hear the true discoverer….the next first author you will read was a drunk who was sleeping while the true worker now languishes in unemployment.

  16. Thomas Lumley says:

    Swiss statistician Stephen Senn has a very thorough one:
    http://www.senns.demon.co.uk/Declaration_Interest.htm

  17. yf says:

    Maybe Nature should disclose how much profit they making by spawning out endless sub-journals, by cornering scientific publishing and by catering to scientific vanity.
    Nature publisher is simply a merchant with a sanctimonious flag of science.

  18. Andrew Marshall says:

    To answer Derek’s question: Nature journals *require* a disclosure of competing interests in a competing interest statement published along with the paper, but what goes what goes into that statement is the prerogative of a paper’s authors (using the guidance provided in the policy about what could be included).

    At the end of the day, the goal is transparency. Does an author wish to disclose financial or non-financial issues that could directly undermine, or be perceived to undermine, the objectivity, integrity and value of their research? On a practical level, this usually comes down to one simple question: if this information became public would it embarrass you as an author? It is context for the research, but as is said above, the research stands or falls on the basis of the rigor of the data.

    Another point: There is a key difference in competing interests between Nature journals and PLoS: Nature journal policy doesn’t consider religious or political aspects to fall within its competing interest policy, for some of the reasons described above. There is no rationale for including them in Nature’s natural science journals: as a researcher, your political or religious beliefs are only relevant if you are carrying out research on political or religious beliefs from which you could stand to gain (e.g., reputationally or in terms of influence).

    For a thoughtful discussion of some of the issues arising with non-financial interests see 2016 paper by Bero and Grundy: http://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.2001221

Leave a Reply

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

Time limit is exhausted. Please reload CAPTCHA.