An inquiry by the British pharmaceutical company GlaxoSmithKline (GSK) into allegations of possible data fabrication in a 2010 Nature Medicine paper regarding the role of specialized T cells in autoimmune disease has found what it sees as evidence of misconduct. Concerns regarding the paper surfaced last week, when news sources reported that the company had begun investigating the research conducted for the study at a GSK lab in Shanghai.

The paper, led by Jingwu Zhang at the GlaxoSmithKline Research and Development Center’s department of neuroimmunology in Shanghai, originally claimed to have found data suggesting that the signaling molecule interleukin-7 caused a subset of T cells known as T helper 17 (T_H17) cells taken from people with multiple sclerosis to multiply. The finding complemented other research in the field suggesting that genetic differences in the cell receptor for interleukin-7 might put some individuals at risk for developing multiple sclerosis—an autoimmune disease thought to involve helper T cells.

Continue reading →

Pretty busy week over at the JAMA offices. First came the report that one of its editors had called a whistleblower a “”https://blogs.wsj.com/health/2009/03/13/jama-editor-calls-critic-a-nobody-and-a-nothing/“>nobody and a nothing”, report that was accompanied by a pretty long series of comments from outraged readers.

Then came the journal’s decision to modify its policy on conflicts of interest. Crucially, the new policy states that “The person bringing the allegation will be specifically informed that he/she should not reveal this information to third parties or the media while the investigation is under way, will be informed about progress of the investigation, upon request, as appropriate, and will be notified when the investigation is completed.”

Ha! I’m sure that those New York Times and Wall Street Journal reporters will be delighted to hold on to their stories before breaking the news that a fresh conflict-of-interest case has come to light. I’m also sure that next time you discover an unreported conflict, you will first inform the journal and wait as long as needed for it to take remedial action, instead of bringing the conflict to the attention of the author’s institution or funding body — what authority do these other people have, anyway?

Not surprisingly, several media outlets have already put their own spin on the way they are reporting this policy change, and they don’t seem impressed by it.

No-one would deny that JAMA has been a leader in raising awareness about conflicts of interest, discussing them perhaps to the point of eliciting a certain desensitization — is anyone surprised when the journal expresses, yet again, the view that conflicts of interest should not be tolerated? Alas, despite its track record, the events of the past week undermine the credibility of the journal’s position on this front.

To my mind, the way in which this whole controversy has escalated is related, in no small measure, to the overzealous way in which JAMA has always decried conflicts of interest. In other words, the tough line that JAMA has taken against conflicts of interest makes the journal much more susceptible to embarrassment when one emerges. Or as the saying goes, the higher they climb, the harder they fall. The latest policy change would seem to be saying that it’s always possible to climb a little higher.

Photo by SparkyLeigh via Flickr

This story in The New York Times got me thinking about how similar high-end restaurants and scientific journals have come to be of late.

Photo Illustration by Tony Cenicola/The New York Times

The article reports that expensive restaurants are no longer playing hard to get and have decided to offer great deals in order to attract costumers. I seem to recall that I read a similar story about British restaurants, but cannot find the link. In any case, the reason why I say that this looks a lot like what’s happening with scientific journals is that it seems that publications are doing everything they can to attract potential authors. For example, according to this blog entry at The Scientist, the Journal of Biology gives authors the option of asking the journal to publish their revised paper without asking the original reviewers to comment on the suitability of the revisions made in response to their critiques.

It seems that the editors of the journal will “carefully scrutinize revised manuscripts,” and if the authors addressed “substantive issues,” the journal will publish the article with an accompanying editorial in which any problems with the paper will be flagged. Sure, authors may be happy with this arrangement, but what about the reviewers? I don’t know about you but, if I were asked to review a paper for this journal, I’m not sure I would be very keen on lending a hand if I won’t have a chance to engage in a dialogue with the authors.

In another example, one of my colleagues at NPG was telling me that a relatively visible cell biology journal has this fast-track system in which members of the editorial board internally referee a paper in less than 2 weeks, only asking for essential controls. Not surprisingly, people in a hurry love this ‘rapid communication’ system. After all, why bother with further experiments to bolster an author’s conclusions?

Then there’s the journal that’s redifining what it means to publish an article — PLoS One. In this case, the only thing that matters is that the paper be technically sound to merit publication. It doesn’t matter if it’s an incremental advance or something not particularly new. As long as the experiments were properly done, the paper will be published. This is actually a very clever model, and I strongly suspect that it will turn on the heat on a lot of specialized journals that publish very thin slices of the scientific salami.

Think about it: if you’re a neuroscientist and your paper didn’t make it in Nature, Science, Nature Neuroscience and Neuron, how much further down the pecking order will you go before you stop caring? The Journal of Neuroscience is a very decent journal, and many of us would still be OK with a paper there. Some of us may go one notch below but, really, very quickly you will want to see the back of that study and just have it published anywhere. PLoS One is therefore an excellent option if your paper didn’t make it into one of the vanity journals, as it will be very visible and freely accessible. My prediction is that very soon this journal will start taking a lot of business from the more specialized journals in every discipline.

There is a problem for the vanity journals, though. If people can publish their work in a decent place like PLoS One, the reputation of which is steadily growing, they will be less inclined to do the hard experiment that will get them a high-profile paper in a vanity journal. This is, of course, bad news for my journal and other highly visible titles. But more worryingly, it might be a bit of a problem for the advancement of science in general, as it isn’t hard to imagine that many scientists may shift into a “complacent mode” in which they cease to ask their staff and themselves to go that extra mile that will turn their study into something really satisfactory. In other words, I can imagine them thinking “why should I do all those experiments that the Nature Medicine referees asked for when I could immediately go to PLoS One and have this part of the story out?”

Don’t get me wrong, though. I don’t mean to insult PLoS One, which strikes me as a legitimate option to disseminate your work. Here I’m trying to make a broader point about the effect that shifting publication standards can have on science at large. In this regard, it may be illustrative to recall the example of PNAS, a journal that, in its heyday, was regarded as a very high-profile publication. I’ve heard many people (including some members of the PNAS editorial board) complain about the fact that members of the National Academy of Science get to publish their work very quickly, after a not-so-stringent peer-review process. I think it’s fair to say that PNAS doesn’t carry any more the weight that it used to carry, but it’s also true that its club-style approach to accepting papers hasn’t been beneficial for the publishing community or for science in general.

The push for attracting papers seems to be so hard that it’s also beginning to affect the vanity journals. Cell, for example, just published this editorial in which Emilie Marcus states that “While some may think the work of an editor is mainly to reject papers, we have found that to achieve our vision for the journal the most important task for an editor is to be an enthusiastic advocate for science and to actively define what is interesting and important to publish—in essence to accept papers.” So, in other words, if you send your paper to Cell you will find an advocate of your science who will try to work with you in order to get the paper where it needs to get.

Emilie is right in that those papers that are potentially interesting but somewhat premature are to be nurtured, and this is something that editors must always try to do — at Nature Medicine we certainly do so. What she fails to mention is that those potentially great papers are so infrequent that, alas, the vanity journals will continue churning out many more rejection letters than letters of encouragement. Be that as it may, as a strategy to get people to submit to their journal, I’m sure the Cell editorial will be very effective.

Even our firm is beginning to experiment with new ways to make a rejection letter from a Nature-branded journal less painful. I don’t think I’m at liberty to discuss the plan in detail, but it is consistent with this global strategy of working in favor of the author, as opposed to asking them to do the hard experiment.

It’s difficult to predict where this whole trend is going to end but, just in case, I’m asking our art editor to print a couple of poster boards like those that top chefs Mario Batali, Sirio Maccioni and Jean-Georges Vongerichten are wearing in the picture above. My plan is to carry the boards with me at every scientific meeting I go to, hoping to attract one or two submissions per trip. You won’t believe our deals — I guarantee it!

A dear friend of mine sent me a link to this page, which shows the “h indices” of what the author of the page refers to the “best Spanish scientists”. The page is a bit difficult to navigate if you don’t know Spanish, but it doesn’t matter; I’m sure that if you have the time and inclination, you will find a similar page in your language and for the nationality of your choice.

The reason for bringing it up has to do with the raison d’etre of the h index — to quantify an individual’s scientific research output. The h index was originally introduced by J. E. Hirsch, from UCSD, in this paper and, briefly, his proposal was that a scientist has an index of h if h of his/her papers have at least h citations each, and the rest of his/her papers have no more than h citations each. In his paper, Hirsch argues why this measure is preferable to other criteria, and ends up suggesting that “this index may provide a useful yardstick to compare different individuals competing for the same resource when an important evaluation criterion is scientific achievement, in an unbiased way”.

I don’t know how many people have bought into this index, but needless to say, as any of these metrics, it has limitations. For example, if you’re the technician of a lab that has a bunch of highly cited papers and you’re always including in the middle of a long list of authors, does your massive h index turn you into one of those “best scientists”?

In any case, its limitations notwithstanding, I thought I would share it in order to stimulate our unsatiable appetite for ways to measure the quality of what we publish. Ready to go check if yours is larger than your neighbor’s?

Image by Brett L.