At 200 years, fortress of medical research confronts the Information Age, Part II

Correction

A quote was wrongly attributed in an earlier version of this story, and the story has been corrected. Tom Easley, publisher and managing director of the NEJM, said the following when asked about an $88 million estimate of its profits that appeared in Boston Magazine in 2005: “It’s not off by a factor of 10.”

Part Two

During the past several decades, medical journals have come under fire for exerting too much control over scientific information.

A new generation of publishers and researchers are challenging the traditional model of highly selective journals that charge hefty subscription fees. Instead, they say, their priorities are to get research out to the public as quickly as possible — and to provide it for free. Some are even bypassing the journals altogether by posting preliminary results online.

“The physical, technological constraints that led to that model are now gone,” said Michael Eisen, who founded the nonprofit publisher Public Library of Science in 2001. The Internet and the demand for instant information, he said, means more scientific research should be easily available.

The New England Journal of Medicine’s current editor Dr. Jeffrey Drazen doesn’t see it that way. Sure, “information and the amount of information is growing,” he said in a recent interview. But he thinks the journal’s publishing model should largely remain the same.

“We’re still being very select about what we choose to publish. And I think that that’s the right thing to do,” Drazen said. “To go through, sift through mountains of information, to find things which we think are going to make a difference to clinicians, so it can make a difference in patients’ lives. That was the challenge of my forbears in 1812 (when the journal was founded) … and it’s my challenge now.”

Public Library of Science is perhaps one of the biggest challengers of the New England Journal’s publishing model. One of its eight journals, called PLOS One, accepts 65 percent of the 20,000 submissions it receives each year. That’s because it rates submissions based on the scientific validity of the research rather than the often-subjective measure of an article’s impact in a field.

The idea is to prevent good research from languishing in the editorial review process of several journals simply because it isn’t considered high-impact enough for a “top-tier” journal, said Eisen.

He gave the example of a gastroenterologist who does research on Crohn’s disease. The New England Journal might reject such a paper simply because it’s too specialized. So then it has to get submitted to a second journal, and maybe even a third.

“Look, this paper might not be important to every doctor, but if you’re a gastroenterologist who treats Crohn’s disease, this paper is really important,” Eisen said. “And that paper might not end up in the New England Journal; now it might end up in some gastroenterology journal, but there’s no reason for that paper to have to have been reviewed first by the New England Journal, and then by some other journal, and maybe by a third journal before it gets published.”

The New England Journal’s brand — as with any top-tier journal — lies in its extreme selectivity. It publishes about 5 percent of the 10,000 or so submissions it receives every year.

But the new “mega-journal” approach is growing. The 172-year-old British Medical Journal, for example, has created BMJ Open, which had an article acceptance rate of 58 percent in 2011.

The creators of PLOS also challenge traditional journals’ restricted access models by posting all material online for free. Eisen said such “open access” policies are more fair to the public, which funds most science and medical research anyway. A significant portion of research is funded by the National Institutes of Health, for example, which received nearly $32 billion in taxpayer money in the 2011 fiscal year.

Rather than charging for subscriptions, PLOS charges submitting authors $1,350 to $3,000 per accepted article, giving exceptions to those who can’t afford it. The publisher collected $12 million in such fees in 2010 and waived or discounted fees for 11 percent of submitters. PLOS turned its first profit that year, of $3 million.

NEJM editors say such a model would fail to support the intense editing and peer review of each article they publish.

“When you buy a journal, it’s not the physical thing,” Drazen said. “Rather, you’re paying the editors to sift through thousands of submissions and bring you the few hundred they think are important. We pay for a lot of stuff you never see.”

He added that the journal’s archives are free to the public online, and all research becomes available to everyone six months after it’s first published.

The NEJM won’t disclose its profits, but when asked if a Boston Magazine estimate of $88 million in 2005 was accurate, Tom Easley, publisher and managing director, said, “It’s not off by a factor of 10.”

The journal’s owner, the Massachusetts Medical Society, made $70 million in publishing revenue and $20 million in advertising revenue in 2009, according to its 990 tax form.

“I don’t begrudge them the right to make a profit on a product,” said Eisen. “It’s the fact that they have a subscription-based business that is so lucrative that makes them, in my mind, unwilling to do the right thing for science and for medicine.”

Despite the resistance of some top-tier journals to the open-access model, Eisen thinks scientific publishing is moving in that direction. Three major research funders — the Howard Hughes Medical Institute, the Max Planck Institute and the Wellcome Trust — recently announced their intentions to fund a new online journal known as eLife, which will be freely available to the public. Instead of relying on full-time editors, the journal has tapped well-known academic researchers to volunteer their time.

The international company Elsevier, which publishes more than 2,600 academic journals, has been boycotted by nearly 10,000 researchers — including dozens at Yale and the University of Connecticut — who say it  charges exorbitantly high subscription fees and prevents the free exchange of knowledge.

A move toward ‘open science’

But beyond calls for “open access,” another trend in the world of research may be even more of a threat to the New England Journal. It’s a movement known as “open science.” In such a model, scientists simply post the results of their research on websites as it’s in process, receiving instant feedback from their colleagues.

Ivan Oransky considers this movement the best reflection of how science works. Executive editor of Reuters Health, Oransky runs the blog EmbargoWatch, which examines how journals control the flow of scientific information.

“Science doesn’t work every week. It doesn’t happen in these punctuated episodes once a week or once a month when the journal comes out,” said Oransky. “We should stop fetishizing the scientific paper, because that’s what’s giving journals right now all their control and I think they’re holding onto that in a defensive way.”

Today, “open science” models are used only by a small subset of researchers, who are generally physicists. Most scientists can’t post their work in the public domain before submitting to a journal, or the journal won’t publish it. That idea was pioneered by the New England Journal’s editor Franz Ingelfinger in 1969, and it’s known as the “Ingelfinger” rule. Since then most major journals have adopted some form of it.

The Ingelfinger rule has been criticized by journalists and researchers who say they can’t discuss their work freely.

“People will show slides at a conference with data, but then if you go up afterwards and ask for a copy of them they’re reluctant to give it to you,” said Gideon Gil, health/science editor at the Boston Globe. “Our role is to tell the public what is happening of importance in medical research, and part of that is to hold people accountable for how they’re using government money, and I think it becomes more difficult for us to do that.”

Drazen said researchers are free to talk about their work, but it should be publicized by the main venue it appears in — the journals.

“We have our own internal rules, and if you don’t like it, you can still publish yourself elsewhere,” Drazen said. “But if you want to publish with us, we want primacy. We don’t want you selling your snake oil or whatever it is, before we have a chance to look at it.”

But the Open Science model is growing. Sites like OpenWetWare, which began in 2005, allow registered scientists to share their work before it comes out in a journal. The idea of unedited information out in the public domain makes the NEJM’s oldest living former editor, Arnold Relman, very nervous.

“That’s a double-edged sword. It may be … the best way to go for research that doesn’t involve human welfare and medicine and medical care,” Relman said. But when it comes to medical information, “without careful peer review and without careful editing, you have chaos.”

Drazen, too, spoke of “a different standard” for medical research. “When you’re deciding something that could make the difference between life or death for a patient, you want to have a different standard.”

But Dr. William Summers, a professor of History of Science and Medicine at Yale University, said that kind of attitude is problematic. The real problem, he said, is that the public puts too much stock in the results of individual experiments that get published in scientific journals.

“People want to believe,” Summers said. “They want to trust some authorities for these decisions that they know are complicated.

“And yet they have to understand that science is a provisional kind of activity, and it’s based on the best evidence we have now, but that may change … and it’s not a fault of science.

“(T)hat’s the way knowledge is generated.”

Part I of this series is available here.

This story by Neena Satija is part of a reporting partnership with WNPR. The audio version for this story — and the audio script — will be available Wednesday at 10 a.m. at yourpublicmedia.org.

The Part I audio script is available here.