NEJM Editor Backtracks on Data-Sharing 'Parasites' Editorial

Marcia Frellick

January 26, 2016

Four days after an editorial by Jeffrey Drazen, MD, editor-in-chief of the New England Journal of Medicine (NEJM), suggested a plan to allow data sharing was flawed because it could breed research "parasites," Dr Drazen issued a new letter clarifying the NEJM's position.

In the letter, published online January 25, Dr Drazen writes: "[W]hen appropriate systems are in place, we will require a commitment from authors to make available the data that underlie the reported results of their work within 6 months after we publish them."

This time, he did not include the word "parasites," saying only that in talking with clinical researchers about the proposal by the International Committee of Medical Journal Editors (ICMJE), "Some of them spoke pejoratively in describing data scientists who analyze the data of others."

He added that researchers who analyze others' data "can substantially improve human health."

Vinay K. Prasad, MD, MPH, a hematologist-oncologist and assistant professor of medicine at the Oregon Health and Sciences University in Portland, called it a "half-hearted statement" that was short of an apology, but that endorses what he calls only a baby step by the ICMJE.

Medscape Medical News previously reported that ICMJE has called for authors to share deidentified individual patient data (IPD) underlying their results no later than 6 months after publication as a condition of their study submission.

Dr Prasad pointed to the phrase "underpinning the results and analysis" published, and said that wording obligates researchers only to hand over the data that support the conclusion they found, not the data that could lead to other findings.

However, he does not think that is enough, he told Medscape Medical News. When a trial involves human beings who have taken a risk by being subjects, he said, a researcher should have to release all of the data: "[W]hat you think is important and what you think is not important."

Original Editorial Criticized Proposal

The earlier editorial by Dr Drazen and NEJM Deputy Editor Dan Longo, MD, concluded that although clinical trial data sharing is a good and noble idea, there is considerable concern about the details as proposed by ICMJE.

Among Dr Drazen's and Dr Longo's main problems with the proposal by the ICMJE is that someone not involved in the original data collection who wants to use the results may misinterpret why the original parameters were chosen.

They also note, "There is concern among some front-line researchers that the system will be taken over by what some researchers have characterized as 'research parasites,' " either using others' research for their own end or using the data to disprove the original hypothesis.

They add that data sharing "should happen symbiotically, not parasitically," and that researchers should identify collaborators who can work with them to further test the hypothesis and have a system of shared authorship.

Dr Prasad said, "History will show this is probably the worst editorial written this year in any journal."

He argued that using a researcher's data and testing them in a different way from a different perspective is not a threat, but precisely what science needs. Other researchers looking at the data may be able to synthesize data from several studies and draw out themes. Reanalysis may also yield important findings that can be practice changing.

"We had a reanalysis of Tamiflu [Genentech], which was once hailed as a panacea for the flu, and now it's been called almost no better than Tylenol [McNEIL-PPC, Inc]," he said. (The reanalysis of oseltamivir published in the BMJ has been referenced in Medscape Medical News.)

He pointed out, however, that any new interpretation of a researcher's data would still have to undergo peer review.

Dr Drazen's editorial sparked passionate replies on social media and on the ICMJE website and inspired hashtags such as #ResearchParasites.

6-Month Rule Too Long or Too Short?

Some commented on the ICMJE website and social media that they take issue with the 6-month postpublication timetable, arguing either that it is too long or that it is too short.

Michael Hoffman, PhD, scientist at Princess Margaret Cancer Centre and assistant professor at University of Toronto, Ontario, Canada, told Medscape Medical News that "allowing data embargoes for 6 months after publication is irresponsible."

"Researchers can fully exploit data they collect until publication. After that, it must be available for the rest of the scientific community to review and evaluate."

Science is a team effort, he says, "but Drazen wants to give extra privileges to one part of the team: clinical trialists such as himself. It's notable that he doesn't suggest gift authorships for those who write the research software or provide the genomic resources critical for many studies."

Elizabeth Wager, PhD, from the University of Split School of Medicine in Croatia, writes in the ICMJE comments section that although she agrees in general with the idea of data sharing, data should be turned over more quickly: at time of submission.

"While I accept that preparing data for sharing can be time-consuming, I worry that journals will not check whether authors have made the data available within this timeframe. It would be much more effective (although tougher) to require it at the time of publication, since this could easily be checked by the journal," she writes.

Some Say Proposal Is Short-Sighted

Others argue that those in countries or institutions with fewer resources run the risk of their postpublication studies being "scooped" because they are not able to complete them as fast as those with more resources.

David Saunders from the Armed Forces Research Institute of Medical Sciences, writes that the ICMJE proposal ignores the need for financial compensation.

"I do not agree that this is something that the ICMJE can or should require.... A typical clinical trial costs millions-100s of millions [of dollars] to conduct," he writes. "Requiring investigators to simply give this information away to the general public, and particularly academic competitors could have significant 2nd and 3rd order consequences that must carefully be considered.... A waiver of all publication fees would seem to be a minimal requirement if this were to be enforced."

He said the requirement to turn over data adds to the growing onerous demands on researchers by regulatory authorities.

"I suspect this would drive at least some authors to seek out new ways of publishing information that 'cuts out the middle man' to leverage the full value of the data and recoup the costs,'" he writes.

The ICMJE will consider feedback before it adopts the new rules. Comments are welcome on the ICMJE website through April 18. New requirements would start with clinical trials that begin to enroll participants beginning 1 year after adoption of the new policy.

Dr Drazen is the editor-in-chief of the New England Journal of Medicine. No other financial relationships were disclosed.

N Engl J Med. 2016;374:276-277. Longo and Drazen editorial full text. Published online January 25, 2016. Drazen letter full text


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.