The JAMA and NEJM Rulings and Their Impact on the Sanctity of Confidential Peer Review

Abstract

A pair of US District Court rulings earlier this year has strengthened the ability of scientific journals to protect the confidentiality of their peer review process. The rulings, which essentially quashed subpoenas Pfizer Inc. issued to a number of journals for a raft of confidential documents, are consistent with past court findings in favor of preserving the confidentiality of peer review. Moreover, they seem to suggest that courts have come to view scientific journals as a segment of the media, and therefore have conferred upon the journals and their contributors some of the same protections given to journalists. The editors of academic journals had closely watched these cases because they were the most significant challenge to the sanctity of peer review in recent times. A defendant in more than 3,000 lawsuits that allege the company marketed Cox-2 inhibitor drugs celecoxib and valdecoxib as likely to provide pain relief without adverse complications, Pfizer sought records from the journals that might contain “exonerating data” in 2007. Journal editors suggested the subpoenas were simply a fishing expedition, as the drug company cast about for anything it could use in its defense. Among the journals that received subpoenas were Journal of the American Medical Association, Archives of Internal Medicine and the New England Journal of Medicine. The journals fought the subpoenas, and in January Pfizer sought court opinions to compel the journals to release the confidential information. Pfizer sought vast disclosures from each journal, including the identity of peer reviewers, comments by and among peer reviewers, and all documents pertaining to the decisions of editors whether to accept or reject submitted papers. Particularly galling to journal editors was the following assertion, made in Pfizer's motion to compel the New England Journal of Medicine to release its documents, which was filed with the US District Court for the district of Massachusetts last January: “The public has no interest in protecting the editorial process of a scientific journal, particularly not when doing so prevents a defendant from access to potentially exonerating evidence. It is unreasonable to conclude that scientists and academics will stop submitting manuscripts to NEJM if it complies with this subpoena.” Yet journal editors weren't arguing that scientists would stop submitting papers if peer review comments were publicly released. Instead they contended that the peer review process itself would grind to a halt for a lack of peer reviewers. And, the journal editors said, the breakdown of peer review would, by itself, be as damaging to the public's interest as a lack of manuscripts. In an affidavit, Dr. Jeffrey M. Drazen, editor of the New England Journal of Medicine, said only about two-thirds of experts approached to engage in peer review agree to do so. By opening up the possibility that their work could potentially be made public, and perhaps still more importantly land them in litigation, fewer still scientists would agree to peer review, Drazen argued. Moreover, he said, ending confidentiality in peer review would lead to less frank comments and impede the scientific process. “Having spent many years intimately involved in the peer review process at the NEJM, it is clear to me that those comments which the peer reviewer knows are not to be shared directly with the author are much more frank in discussing the strengths and weaknesses of a manuscript,” Drazen argued. “This is not to say the comments that the reviewer knows may be shared with authors are any less valuable, only that the reviewers are more tempered in their assessment when they know that their comments may be disseminated beyond the editorial office.” The district courts in Chicago, for Journal of the American Medical Association and Archives of Internal Medicine, Boston, for the New England Journal of Medicine, had to weigh the relevance of the materials sought by Pfizer in its legal defense against the journals' interests in protecting peer review. For the journals, the cases amounted to nothing short of total violation of peer review's sanctity. These were monumental cases. The Chicago court ruled first, with Magistrate Judge Arlander Keys finding that Pfizer could obtain the information it sought from the published articles themselves, rather than dredging it up from the confidential material held by Journal of the American Medical Association and Archives of Internal Medicine. Keys wrote: “Especially given the strong policy behind preserving confidentiality in the peer review process, the Court finds any probative value would be outweighed by the burden imposed on the journals in invading the sanctity of that process.” Journal of the American Medical Association editor Dr. Catherine DeAngelis said she was “delighted” with the ruling. Magistrate Judge Leo Sorokin, of the Massachusetts court, was more cautious in his ruling on the New England Journal of Medicine case in late March, saying Pfizer made a “prima facie showing of relevance.” Still, he sided with the academic journal's interests: “On the other hand, the NEJM's interest in maintaining the confidentiality of the peer review process is a very significant one,” Sorokin wrote. “We are pleased that the confidentiality of the peer review process remains intact,” said Karen Pedersen, a spokeswoman for the New England Journal of Medicine, following the ruling. Significantly for all academic journals, Sorokin found that the New England Journal of Medicine was “entitled to a level of protection commensurate with that afforded journalists.” Like journalists, Sorokin concluded, academics were information gatherers and disseminators. While such protections may sound lofty, there are considerable limits on the legal comforts afforded to journalists, said Mathew Bunker, a professor of journalism at the University of Alabama who has worked as both a journalist and a lawyer. Bunker cited a recent string of high-profile cases–most notably the Valerie Plame-Scooter Libby affair in which New York Times reporter Judith Miller was jailed for 85 days – in which federal courts have compelled journalists to reveal their sources. “Journalists have been on a losing end recently,” Bunker said. “In the case of the New England Journal of Medicine, the court made a very solid decision. It's unfortunate that these interests aren't always recognized, especially the importance of protecting sources behind the scenes in the gathering of important information.” Bunker also noted Sorokin wasn't explicit about the source of protection for the New England Journal of Medicine in his ruling, making his decision instead based upon what was good for public policy, namely the importance of protecting the privacy of peer review. Despite the judge's ruling, the former journalist said it's also unlikely that academic journals would enjoy the protections of shield laws, which exist in 32 states. These laws, which vary from state-to-state as there is no federal legislation, allow journalists to refuse to testify and reveal sources in certain types of proceedings. Yet in many cases, Bunker said, the state shield laws specify the protected media – such as newspapers, radio and television – thereby excluding academic journals. In previous legal precedents involving peer review document subpoenas, journals have generally won challenges to their rights to protect the privacy of peer reviewers and their reviews. Depending upon the circumstances of a case, a journal might be protected by a journalist's privilege, a scholar's privilege or the First Amendment. Such cases aren't all that common. In his more than 25 years with Annals of Emergency Medicine, Dr. Michael Callaham, this journal's editor, said Annals of Emergency Medicine has never had any legal issue where it was requested to reveal peer reviewer identity. But it has happened enough to make journal editors and peer reviewers wary of courts. One of the most famous cases occurred in the early 1990s when Jean Cukier and his co-authors submitted a manuscript to the Journal of the American Medical Association, and Cukier included a statement that he had no financial interest in the publication. The journal editor later sent Cukier and a coauthor a letter informing them that someone had provided information about a related financial interest Cukier held. Cukier denied the financial interest, but Journal of the American Medical Association declined to publish the article. Cukier sued in 1994 to learn the identity of the person who questioned his integrity, and the Journal of the American Medical Association defended itself by citing the Illinois Reporter's Privilege Act. The court sided with the publication, holding that the journal could have learned the information from an undisclosed source during the “news gathering” process, and that the Journal of the American Medical Association met the qualifications of a “news medium.” Another case came later in the decade, when a series of women were falsely diagnosed with carcinomas after assays found high levels of the hormone hCG. These tests led doctors to search for tumors using imaging techniques, and then surgery, and in some cases women had their ovaries removed. Researchers first reported these false positive tests in the journal Clinical Chemistry in 1997. The assay's manufacturer, Abbott Laboratories, was subsequently sued. During the case the journal's editor, Dr. David Bruns, a professor of pathology at The University of Virginia, received a subpoena for all peer review documents associated with the published article. Bruns hired a lawyer who asserted the documents were protected by “among other privileges, the newsgathering and editorial privilege, the peer-review analysis privilege, and the self-evaluative privilege.” The court sided with Clinical Chemistry. “Our lawyer, fortunately, made the issue go away,” Bruns said. “I think we could have won – I haven't heard of many cases where a journal has lost an issue like this – but this would have cost of tens of thousands of dollars to litigate in court.” Bruns said lawyers want to identify peer reviewers to put them on a witness stand as well as to essentially get free advice for their case. “Anyone who begins to think about the ramifications of this for a while, you begin to shudder,” he said. The effect of losing the shield of confidentiality surrounding peer review would have an equally negative impact on the quality of science, added Bruns, chief editor of Clinical Chemistry for 17 years. “I think it would have a chilling effect on the willingness of reviewers to state exactly what they think about papers,” he said. “There are some younger editors who think their peer reviewers are perfectly happy to have their names made public. But I've heard too many times from reviewers who are absolutely horrified if their names are accidentally revealed to authors, especially older, influential ones.” The efforts to protect the confidentiality of peer reviewers comes as an effort to move toward open peer review – in which the reviewers' names and reviews are made public – has stalled somewhat. One recent survey, in fact, found a preference among academics for an even more secretive process, double-blind review. The survey of 3,040 academics, 90% of whom had been reviewers, found general satisfaction with the overall peer review process. Commissioned by the Public Research Consortium, and conducted by Mark Ware Consulting, the survey found that 85% of respondents believed peer review greatly helped scientific communication. Some 56% of respondents expressed a preference for double-blind review, compared to 25% for single-blind review, 13% for open review, and 5% for post-publication review. And yet, for all the support of double-blind review, it hasn't gained wide acceptance. Editors of the open-access Public Library of Science recently abandoned double-blind peer review because too few authors requested it, and the authors were too easily identified by reviewers. Open-peer review appeared to be gaining momentum about a decade ago. The British Medical Journal adopted open peer review in 1999, and Nature conducted a 4-month trial of open peer review but chose not to adopt the process, concluding that, “Feedback suggests that there is a marked reluctance among researchers to offer open comments.” Maxine Clark, Publishing Executive Editor for Nature, said, “We are happy with our current peer review system, including our editors' selection of peer reviewers and their management of the peer review process.” Single-blind peer review, then, would seem to be here to stay. The newly reinforced message from the courts to journal editors, peer reviewers, authors and the public which benefits from sound science, is that confidentiality of the process is here to stay as well.