Publishing, and the Subtleties of Conflict of Interest
A conflict of interest (COI) is (very roughly) a situation in which a person has a private or personal interest sufficient to appear to influence his or her professional judgment.
One important category of COI involves scientific (or, more generally, scholarly) publication. It is typically regarded as a COI, for example, for a researcher to publish findings (typically not just reporting, but interpreting data) based on research funded by a drug company whose fortunes may rise or fall based on those findings. (In publishing, COI is typically referred to by means of the more genteel term, “competing interests.”)
Now, it’s crucial to make clear that to be in a COI is not, in and of itself, unethical; what matters is how you deal with it. Most science journals now require that authors disclose “competing interests” (i.e., primarily financial interests) that might reasonably be suspected of biasing the things they write. See, for example, the Nature journals’ competing financial interests policy.
Requiring authors to declare competing interests is a good idea, but it’s unclear just how it helps readers. The theory, in principle, is that if I, as reader, know that an author’s work was funded by a drug company, I can attenuate my level of credence suitably. OK, but just how do I do that?
Here’s an example. This story in the journal, Nature, garnered a lot of press last week: Towards responsible use of cognitive-enhancing drugs by the healthy. I agree with a lot of what the authors say. But it’s a useful example of the complexity of COI. Note that at the end of the article, there’s a notice saying that two of the authors declare “competing financial interests”. Basically, two of the article’s 7 authors work as consultants for pharmaceutical companies.
How should that revelation affect our evaluation of the article? I teach my first-year Critical Thinking students that you can’t discredit an argument simply by telling me about the person who wrote it: that’s called an ad hominem argument, and those are generally considered fallacious. (Crude example: if my argument is that All A’s are B’s, that all B’s are C’s, and that therefore all A’s are C’s, it doesn’t matter who I am or who funds my research, because that’s a valid deductive argument. Period.)
But not all arguments are so straightforward. The Nature article cited above, for instance, puts forward a multi-pronged argument, citing various reasons that the authors take, in the aggregate, to lead to the conclusion that we ought to adopt a more permissive attitude towards drugs that enhance brain function. And in putting forward that argument, the authors necessarily faced choices about what words to use, what analogies to employ, and what rhetorical devices to rely on. And it’s at least conceivable that those choices could be swayed by interests other than the pursuit of truth and the public good.
As I said above, I think the Nature article puts forward a pretty good argument. But I’m more comfortable with that assessment, knowing that I’ve thought carefully about the way the authors’ choice of words and analogies might (or might not) have been affected by the funding of their research.