Funder-supported journals are just coming onto the scene. One of the most prominent — eLife — is billed as “a unique collaboration between funders and practitioners of research to communicate influential discoveries in the life and biomedical sciences in the most effective way.”

The word “collaboration” may be more loaded than intended.

Journals are supposed to provide disinterested evaluation of research papers in order to validate, filtrate, and designate them properly in the scholarly ecosystem, a process that often includes rejection or deep revision. Typically, funders aren’t allowed to influence publication decisions around research they’ve funded. In fact, even the appearance of funder influence can be deemed a conflict of interest (COI).

Editors who assume the helm of a journal typically have to rid themselves of any competing interests. One editor I worked with during his early tenure recused himself from evaluating papers from his past funders for 18 months after taking the helm. While he could probably have provided disinterested editorial oversight, the perception that he would be conflicted was too risky, so recusal from certain decisions was adopted for a year and a half.

Since eLife is aiming for biomedicine and for the top tier, it seems appropriate to refer to the International Committee of Medical Journal Editors (ICMJE) definition of COI:

Conflict of interest exists when an author (or the author’s institution), reviewer, or editor has financial or personal relationships that inappropriately influence (bias) his or her actions. . . . Financial relationships (such as employment, consultancies, stock ownership, honoraria, and paid expert testimony) are the most easily identifiable conflicts of interest and the most likely to undermine the credibility of the journal, the authors, and of science itself.

I was recently in Frankfurt, where Robert Kiley, a representative of the Wellcome Trust, gave a talk promoting eLife. It’s clearly a Wellcome-funded initiative. The editors are employees of eLife, as are the editorial staff. The organization is set up as a separate 501(c)3 registered in Delaware in the US, with another registration in the UK. Delaware is one of the two easiest states in the US to create a legal business entity (the other is Nevada). In addition to tax advantages, one major advantage of setting up a corporation in Delaware is that the directors or officers don’t need to live in Delaware.

So we have a clear ownership and funding arrangement — eLife receives funding from Wellcome, HHMI, and Max Planck; eLife hires editors and editorial staff with this funding; and eLife currently allows itself to review and will apparently allow itself to publish papers funded by its owners. In fact, one major motivation of starting eLife was apparently to make it more certain that papers funded by Wellcome and HHMI get published. As Kiley put it, “productivity” has gained preeminence as their goal. Publication is key to how they measure the productivity of their funding choices, and this productivity is used to get further funding.

The question this all raises is, Can the editors of eLife work for eLife while also evaluating and editing papers funded by Wellcome and HHMI?

They will certainly say that they can, but in the world of COI, it’s not for them to judge. Individuals or entities involved can’t assess their own COI exposures. External, independent people and organizations should do this. And this is exactly the problem — they are not in a position to judge.

A further reading of the ICMJE guidelines particular to editors and editorial staff illuminates the problem more clearly:

Editors who make final decisions about manuscripts must have no personal, professional, or financial involvement in any of the issues they might judge. Other members of the editorial staff, if they participate in editorial decisions, must provide editors with a current description of their financial interests (as they might relate to editorial judgments) and recuse themselves from any decisions in which a conflict of interest exists.

According to this, the editors of eLife should be recusing themselves from evaluating any papers funded by their employers. After all, they have a financial interest in publishing them — they are employed by the funders, and have a professional and financial interest in the outcome.

I was speaking about these issues with an editor from a high-profile journal. It became clear that this arrangement’s inherent COI will likely put authors in a tough spot:

1. I received funding from Wellcome or HHMI.
2. I did a study, and it turned out well.
3. On the one hand, it makes a lot of sense to get the paper published in a very high-impact and prominent journal in the field I work in. On the other hand, HHMI and Wellcome have established a mega-journal of their own. Should I submit it to their journal? Will they punish me if I don’t? Or should I submit it to the top journal? Which path is better for my career?

This is not an easy dilemma for a researcher or research team to solve. I’d argue that it’s a position they shouldn’t be in. It’s directly derivative of a COI problem between the funders and eLife, a problem that exists in any version of the model behind eLife’s funding.

Of course, there’s another scenario for this thinking, one we can’t ignore:

1. I received funding from Wellcome or HHMI.
2. I did a study, and the results were weak.
3. I can probably get it published pretty easily in my funders’ mega-journal, so I’ll do that.

Central to all of this is the issue of aligned interests, and that’s when the COI dimension becomes clearest. It’s well-known that there’s always a temptation and even a tendency for researchers to overstate their results — there are a lot of incentives driving this type of behavior, both intellectual and financial. These same incentives drive funders — they want the research they’ve funded to be as impactful as possible, both to validate their funding decisions and to attract future donations. Neither of these interests are scientific interests — they are personal, professional, and financial interests, which often conflict with scientific interests. Because the interests of authors and funders are so highly aligned, they need to be separated by an independent third party capable of objectively validating the results.

For the editors at eLife, it may not be clear which interests they are serving — author or funder/owner — because these interests are so similar and highly aligned. The authors and funders want as great an impact as possible. There is a significant sunk cost for the researchers — months and years of work, the effort of assembling and analyzing results, and the effort to write the paper. There is a significant sunk cost already from the funder’s perspective, and as the owner of the journal, this may be perceived as another source of bias. In the face of these strong financial, personal, and professional biases, there is an editorial staff employed by the funders themselves.

The typical defense in situations like this is, “It won’t happen here. We’ll regulate ourselves carefully, and monitor what we do.” The problem is that self-regulation doesn’t seem to work. Recently, in the United States, the Food and Drug Administration (FDA) has allowed food companies to regulate themselves and conduct their own inspections. The result? More deaths from infected foods — cantaloupe, spinach, meat, and peanut butter.

David Kessler, former commissioner of the FDA, said this about the conflicts of interest this self-regulation approach has created in the financial markets:

There’s a fundamental conflict. We all know about third-party audit conflicts. We’ve seen it play out in the financial world. You can’t be tied to your auditors. There has to be independence.

Auditors were once “independent” in name only, which led to major problems in the financial markets. Not learning from that over the past decade, financial industry regulations were rolled back so that “market forces” could compete more effectively, with the result being too much risk, too little transparency, too much cronyism, and a worldwide economic meltdown. It’s pretty clear that creating unbroken lines from source to market is a bad idea.

There is a stalking question here, too — Can eLife and journals structured like it be perceived to fairly evaluate research coming from other funders? Why should Moore Foundation researchers believe they’ll get a fair shake from a journal funded by another group of funders? Will commercial funders feel they’ll receive an unbiased review from editorial staff effectively paid by other funders?

Disclosure is one way of dealing with COI — it at least informs readers that a potential conflict exists, and lets the reader judge for themselves whether that conflict is manifest in any way. The disclosure statement for eLife could be something like this, on every article:

Wellcome Trust [or HHMI, or both] funded this research, and also provides funding for the editors and editorial staff of eLife.

OA publishing has some new COI dimensions we need to grapple with. Should OA publishers tell readers who paid the article publication charges (APCs)? It has a bearing on perceptions of the interests behind the publication, so I think it’s a no-brainer — it’s clearly the best high-road action, and it’s very easy to do. Now, we see a larger COI issue arising as funders become publishers, fund research initiatives, and hire editorial staff to evaluate the reports they’ve funded. Should this be allowed? Will authors accept it? Should authors even have to feel incipient pressure from a conflicted business relationship?

To me, this is a clear case of improperly aligned incentives, and therefore should not be allowed to exist in scholarly publishing. We have enough scandals and failings already. A massive COI failing on the scale proposed by the existence of eLife is clearly a step too far.