14

It has been proven that the source of funding can have an influence on the results of a research (funding bias, well documented here and here).

When we check for pieces of research, for example through the Pubmed database, how can we check the source of funding? (And check for possible financial conflict of interest?)

7
  • 3
    I think that your first statement urgently calls for references. As for determining sources of funding, AFAIK, research papers typically include acknowledgement of funding sources and even specific grant references. Many authors also include conflict of interests statements. Oct 27, 2015 at 3:07
  • Are you referring to government funding? Private funding? something else?
    – ff524
    Oct 27, 2015 at 3:20
  • @AleksandrBlekh : Just added some references you asked for. Does this mean that there is no way to verify the funding source of a study, other than look at what is claimed by the authors in the paper ?
    – lapin
    Oct 27, 2015 at 3:50
  • 1
    I asked because it's easier to find out who gets government funding (NIH RePORTER, for example). For private funding, there's no way to check other than what authors themselves report in the paper. (Why should there be a way to check? Who is supposed to be responsible for verifying funding source?)
    – ff524
    Oct 27, 2015 at 4:02
  • 2
    Reputable journals require authors to disclose funding sources and other financial interests, which are then displayed in the article text. See for example here.
    – Dan Romik
    Oct 27, 2015 at 4:55

3 Answers 3

2

Reputable journals might require authors to disclose funding sources, but as noted, this practice, reputable journals notwithstanding, is inconsistent. Beyond this, there are no standards for determining what is a reputable journal.

As to the comment that two citations do not prove the point, the literature is replete with studies demonstrating funding bias; a few examples: Pharmaceutical industry sponsorship and research outcome and quality, Lexchen, Bern et al; Avoiding biasing the conduct and reporting…, Hillman, Eisenberg, et al; Systematic review of empirical evidence of study publication bias, Dwan, Altman et al.

1

I wouldn't say that two studies is "proven," especially when neither of those is empirical: The first one is a literature review, and the second one is a theoretical article. Besides, both of them talk almost exclusively about the funding of medical/pharmaceutical research by pharmaceutical companies, and not about funding agencies as a whole.

That said, journals are supposed to be the gatekeepers of this information. When scientists submit a journal article, most journals these days ask them to disclose a conflict of interest: whether their funding or other personal interests is tied up with the research they're doing. Some journals publish this information publicly for readers to see, and others do not. Some journals require a funding statement - a little paragraph in the footnotes that states where the funding comes from

But this is done inconsistently; there's no standard from journal to journal, and not all journals report the information at all. The one "exception" is NIH funding - because of new NIH regulations, scientists have to disclose their NIH funding in the journal article and link their funding to the articles themselves, so theoretically the public can see where their tax dollars are going. In practice, the process of linking can take several weeks to several months (although this will probably speed up as it becomes more commonplace).

So the answer is - often times the only way to check IS what authors themselves report in the paper, short of contacting the author(s) and asking them yourself.

1
  • I would note that there are many other NIH-type "exceptions" - it's a policy for all UK RCUK-funded grants, for example - but exactly how effective these policies have been is not yet clear.
    – Andrew
    Oct 27, 2015 at 9:26
0

All funds I've been involved in did require thanking the source (and often state the exact grant) in any publications.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .