Researchers Push Back Against Journals’ Demands That Medical Data Be Shared

August 04, 2016

Hundreds of university researchers pushed back on Wednesday against a proposal by editors of the world’s top medical journals to require that data from clinical trials be quickly shared. Instead, the researchers urged terms that would delay the period before data must be disclosed and provide economic and academic compensation to researchers who carry out the trials.

The plan, which was outlined in January by the International Committee of Medical Journal Editors, would require comprehensive data sharing within six months of an article’s publication. But a group of more than 300 researchers, writing in The New England Journal of Medicine, said that proposal was far too hasty.

The researchers noted that journals publish the results of more than 25,000 clinical trials a year. Imposing a firm requirement for sharing the underlying statistical data, they said, could raise "enormous costs" for researchers and lead to misleading or inaccurate analyses.

The group, organized by scientists at McMaster University, in Ontario, and endorsed by 282 investigators in 33 countries, suggested pushing the "exclusivity period" — that is, the time before data must be shared — to at least two years and as many as five. The researchers also proposed requiring independent statisticians to conduct reviews of released data.

Theirs was one of four articles on the topic published on Wednesday by The New England Journal of Medicine. One of the other articles, signed by more than 200 scientists involved in cardiovascular research, also suggested at least a two-year process allowing for independent reviews of requests for access to clinical-trial data.

Money and Academic Credit

The responses to the International Committee of Medical Journal Editors, which represents the world’s most prestigious medical publications, show that, despite years of efforts by many scientists and their funders to encourage data sharing and open access to publicly financed research, persistent hurdles remain.

The medical-journal editors said such data sharing is an ethical imperative that will speed research and raise confidence in medical discoveries. But skeptics have warned that plans for sharing the costs of conducting trials and ensuring the quality and coherence of data should be in place before firm requirements for sharing can be imposed.

Another of the articles appearing on Wednesday in The New England Journal of Medicine was written by Sen. Elizabeth A. Warren, Democrat of Massachusetts and former professor of law at Harvard University, who largely lauded the editors’ proposal as a needed push in a realm where federal attempts have met with mixed results.

The National Institutes of Health already has data-sharing requirements, but Ms. Warren noted that compliance has been spotty. Enforcing those requirements has been difficult, she wrote, in part because of lingering disagreements among policy makers that have left the government without a final set of regulations.

Two key issues needing resolution are how to provide money and academic credit to the researchers who produce the data. If those issues aren’t dealt with, said P.J. Devereaux, a professor of clinical epidemiology and biostatistics who led the group at McMaster University, then journal rules requiring data sharing could backfire, making valuable trials more scarce.

“The academics are probably the slowest to pick up on this, but people feel it's inevitable that we'll get to a place where there'll be an expectation the data are shared.”
Researchers may have themselves to blame for not having driven the process earlier, said Manesh R. Patel, an associate professor of medicine at Duke University who helped organize the response from cardiologists. "The ICMJE is going to force us to have the conversation maybe a little faster than we’ve been having the conversation," Dr. Patel said, referring to the medical-journal editors.

Most researchers seem to recognize the time has come to seriously assess what trade-offs they need to accept to ensure wider sharing of clinical-trial data, said Harlan M. Krumholz, a professor of medicine at Yale University and an author of another of the New England Journal of Medicine articles.

"We’re on the cusp of a real breakthrough, we’ve been making a ton of progress," Dr. Krumholz said. "The academics are probably the slowest to pick up on this, but people feel it’s inevitable that we’ll get to a place where there’ll be an expectation the data are shared and made available, that the investigator doesn’t have sole control over them."

Paul Basken covers university research and its intersection with government policy. He can be found on Twitter @pbasken, or reached by email at