The recent revelations about how eLife and PubMed Central secretly cooperated to give eLife access to US government infrastructure to support its launch, allowed eLife to skip the application and review process for inclusion in PubMed Central and PubMed, and sought to conceal these facts not only from its own advisory committee but from other publishers who inquired raises many specific questions about how PMC is managed and how its conflicts of interest are handled. It is also beginning to raise questions in my mind about the funders associated with eLife and the PMC events, who have been silent on the controversy.

The story has unfolded over many months, and the NIH and NLM have been very aware of it, based on internal emails obtained from a Freedom of Information Act (FOIA) request. Since the first post last October, there were indications within emails from NIH and NLM employees to librarians and others that the NIH was “working on a statement in response,” but no statement has ever been released. When contacted by a journalist in November, David Lipman is quoted as saying, “At this time, we don’t think that engaging Mr. Anderson in a public rebuttal is productive.”

The scandal has only gained momentum and accumulated evidence since then, and the emails and communications obtained earlier this month have been described by professionals in the scientific publishing community as appalling, disgusting, worse than imagined, and astonishing.

While none of the funders involved has responded despite requests for a statement about the controversy, eLife’s Executive Director Mark Patterson — who was the person who requested the favor from PMC in the first place, and who also sought and accepted editorial input from US government employees — did respond to a request last week. Here’s the bulk of his response:

The rest of the email is self-promotional.

There are a number of evasive and disingenuous aspects to Patterson’s response:

• “Open-access publishers are keen to have their content presented on PMC because it is such an effective platform.” Indeed, which is why so many OA publishers are angry that eLife got to cut the line and avoid the normal process. Because PMC helps validate and disseminate OA content, many OA publishers have been working to meet published requirements. This takes time. But Patterson wants to skip over why eLife was allowed to be included in PMC prior to even having the ability to accept manuscripts. Perhaps more pernicious is that Patterson views PMC as a “platform,” like ScribD or Mendeley. That is, as he portrays it in his response, the platform is free to use by anyone. If this were true, we’d be having a very different discussion — that is, where would HighWire and Atypon and Silverchair, and Publishing Technologies and others then stand in relation to PMC? But PMC is not a technology platform everyone can use whenever they want. It is a library resource bound by rules it has promulgated for years. ELife was allowed to skip those rules. Patterson is eliding the facts. I find that to be another blow to credibility.
• “We initiated communication with PMC about this idea early last year, and were included in PMC about 7 months later, once we had met all of the technical requirements.” No, that’s not what the documents show. They show that eLife was given its current entry into the NLM database within seven business days, not seven months. A request to publish eLife articles via PubMed Central was made by Patterson on February 24, 2012, and a record for eLife (#101579614, which remains eLife’s catalog number) was created on March 5, 2012, barely even seven business days later. There is no evidence of eLife even filing an application for inclusion in PMC, and when discussions started, eLife didn’t even have an ISSN or a URL, both requirements for the PMC application process. Internal emails at NLM have managers referring to eLife apologetically as “not your usual piece of cake.”

I emailed Patterson about this directly, and received more obfuscation:

His answer:

I have yet to see any evidence that eLife even applied for inclusion into PubMed Central — it certainly didn’t meet the stated requirements to apply as of February 2012, and was pursuing using PubMed Central as a publishing platform, not applying to be included, in the email Patterson references. Again, Patterson is being evasive. He could have just said, “No, we didn’t ever officially apply.” Instead, he tries to smudge it over.

• “eLife took the unusual step of seeking to make its initial accepted articles available via PMC (and other sites), two months ahead of the launch of the journal website in December. We explored this because it was clear very early in the project that we would have content accepted well ahead of the availability of the journal web site, and we wanted to get this content out as soon as possible. “ Again, Patterson is shading the truth. ELife’s launch date moved a few times during 2012. There was nothing as neat as “two months ahead of launch.” From what I can tell, it was the usual scramble, with the launch date pushed back a number of times. But eLife had PMC’s assurances and help from late-February through December 2012. The truth is that eLife requested special treatment from PMC before eLife could even accept papers. This suggests eLife was presumptuous about the quality of papers it would receive — before it could even process any papers. ELife requested special treatment when its editors were still using Wellcome email addresses (eLife still didn’t have its domain squared away), and with a shifting launch date. Patterson contradicts himself in the same email, stating, “The calibre of the research we have selected for publication, which began to come to light in the summer, has since spoken for itself.” So, Patterson requested early publication via PMC in February 2012, but the quality of the research came to light in the summer? Granted, eLife was learning there was some good research out there early on, but that’s true for any start-up journal, and hardly an excuse to operationalize conflicts of interest in order to gain a market advantage.

Publishers should behave with integrity. These answers from Patterson are not impressive examples of integrity or transparency. They are evasive and self-contradictory.

When it comes to conflicts of interest, a story from US News & World Report about eLife bears some scrutiny. I was interviewed for this story, where I brought up my belief that funders backing a journal on an ongoing basis creates an inherent conflict of interest. This can be contrasted with how PLoS started, where a foundation gave PLoS a few million dollars to start up, and then left them to their business. The eLife situation is not nearly as clean. Its three funders have maintained an ongoing interest, and each funds an editor’s research.

In the US News story, Robert Kiley of Wellcome Trust  is quoted:

Kiley, from the Wellcome Trust [says] the backers are “very comfortable that we have raised proper safeguards against any actual conflict of interest,” he said in reference to Anderson’s claims.

Kiley clearly isn’t aware of what constitutes a conflict of interest or how to handle them. There is no practical separation between “actual” and “perceived” when it comes to conflicts of interest, and for very good reasons. As I quoted in my first post about the conflicts of interest between PMC and Wellcome Trust, from a National Academies book on the topic:

All conflicts of interest involve perceptions or appearances because they are specified from the perspective of people who do not have sufficient information with which to assess the actual motives of a decision maker and the effects of those motives on the decisions themselves. . . . A conflict of interest is a set of circumstances that creates a risk that professional judgment or actions regarding a primary interest will be unduly influenced by a secondary interest.

Yet, Kiley asserts that Wellcome is above this issue, because it will prevent “actual” conflicts of interest. How can we know this? How can we know that the secondary interests of the editors and publishers and eLife — each of whom receives funding from one of the backers, making it in their interest both as employees and as scientists to please Wellcome, HHMI, and Max Planck — are not affecting their decision-making? Which set of interests is actually primary to them?

We’ve already seen conflicts of interest between Wellcome and PMC move from the “perceived” to the “actual” with Wellcome, eLife, and PMC cooperating to give eLife not only inappropriate help launching, but also an unprecedented spotlight at a PMC NAC meeting. Mismanagement of conflicts includes Patterson and David Lipman, Executive Secretary of the PMC NAC, not informing the NAC of their plan to put eLife content on PMC despite the June 2012 meeting having a clear section devoted to a discussion of hosting plans.

Since the revelations of three weeks ago, there has been no refutation and no apparent accountability from any of these large funders — Wellcome has been silent; HHMI has been silent; Max Planck has been silent; and the NIH has been silent. The only statement we have — from eLife — shades the truth, evades responsibility, and seems smug and unaccountable.

ELife has been flip about this all along. Patterson mentioned the first posts from October at a publishers meeting last year, equating them to “marketing” for eLife. Randy Shenkman, editor-in-chief of eLife, was asked about the more recent articles at another publishers meeting this year, and gave an answer that struck many present as glib and dismissive. But here’s what seems beyond dispute:

• eLife was allowed to be included in PubMed Central before it should have even been allowed to apply; in fact, it never officially applied and was accepted before it even met the criteria for application
• eLife was published on PubMed Central for two months, at US taxpayer expense
• eLife requested special treatment, specifically via Mark Patterson, now its Executive Director
• Patterson and Lipman avoided telling the PMC NAC about their arrangement in June 2012, despite the topic of hosting being on the agenda
• eLife and NLM employees worked to eliminate wording from an eLife editorial and an email to PeerJ to conceal the extent of their collusion
• eLife was accepted into PubMed Central without an application and within 6-7 business days
• Wellcome Trust was given an exceptional presence on the PMC NAC (two consecutive terms), and Patterson’s original email requesting special treatment came from his Wellcome email address
• The appearance of conflicts of interest are only the beginning; these conflicts have become real in the case of PubMed Central, Wellcome, and eLife, raising concerns about whether these three entities know how to manage the conflicts of interest inherent in their market positions

I can only imagine what kind of hue and cry would be battering Elsevier or another not-for-profit or commercial publisher if they remained silent in the face of evidence of such conflicts of interest, biased treatment in what is supposed to be an objective system, and overt cronyism. Then, on top of that, eLife is giving responses which amount to brush offs. The Elsevier “fake journals” scandal resulted rapidly in an apology and reassurances from Elsevier. Other journals caught with conflicts of interest on their editorial or executive staff — perceived, not actual — have moved quickly to eliminate the perceived problem and have apologized to the community. ELife seems to feel to compunction in this regard.

All this evasion, double-dealing, and cronyism is starting to make me wonder if eLife has the right people on board. Maybe I’ve been spoiled throughout my career, but I’m accustomed to there being some admirable person present in these situations who takes things seriously and isn’t glib and dismissive of potentially serious issues. I have yet to see this kind of behavior at eLife — either in the Editor-in-Chief or the Executive Director.

More fundamentally, where is the voice of concern or regret coming from Wellcome, HHMI, Max Planck, or the NIH?

The silence is becoming deafening.