Face To Face

As editor of the New England Journal of Medicine for more than a decade, Arnold S. Relman has played a significant role in setting publication standards for scientific journals. He champions the “Ingelfinger rule”promulgated by his predecessor, Franz Ingelfinger, which bars contributors from publicizing their articles before publication in the Journal. He also has strongly supported embargoes that permit reporters to receive advance copies of scientific journals on condition that th

Mar 21, 1988
Tabitha Powledge
As editor of the New England Journal of Medicine for more than a decade, Arnold S. Relman has played a significant role in setting publication standards for scientific journals. He champions the “Ingelfinger rule”promulgated by his predecessor, Franz Ingelfinger, which bars contributors from publicizing their articles before publication in the Journal. He also has strongly supported embargoes that permit reporters to receive advance copies of scientific journals on condition that they withhold any resulting news stories until a specified time.

Such policies have been challenged recently In its January 28 issue, the Journal published results of a clinical trial indicating that male physicians who took an aspirin every other day had fewer heart attacks than colleagues who took a placebo. The Reuters news agency ran a story on the study before the release date; Relman responded by suspending the wire service’s special airmail subscription for six months.

Q: Was there much controversy over your decision to publish the report by a committee of the University of California at San Diego on 137 articles by Robert A. Slutsky?

RELMAN: No, I’m not aware of any. Q: Some of the articles that committee identified as either questionable or fraudulent have still not been retracted, apparently because the journals in question are worried about litigation. Is that fear appropriate?

RELMAN: Fears about litigation are always appropriate in our litigious society. Whether they are justified is another matter.

When the responsible sponsoring institution makes a determination that a manuscript or published report is fraudulent—after a thorough investigation in which the authors have had a full opportunity to defend themselves—then it’s their obligation to so notify the journal. When the journal gets such notification, the editor has an obligation to do what is requested of him—namely, to retract the manuscript.

If the institution does not notify the journal, then the journal should not act unilaterally. But if the institution notifies the journal, or of course if the author himself asks for a retraction, the journal editor is obligated to do that. I can’t imagine the editor being found guilty of misconduct. He is simply doing what his position requires him to do. Furthermore, when questions of fraud are raised, it is the editor’s obligation to see to it that the issue is addressed.

I have on a number of occasions been given information that raised a serious possibility of fraud. As an editor, I have no authority to conduct an investigation myself, nor the resources to do that. But what I have done is go to the sponsoring imtitution and say, this is what I heard. Please investigate it however you see fit, and please let me know what you decide and what you think I ought to do about it.

Q: Are you happy with the kinds of responses you have gotten?

RELMAN: Yes, so far.

Q: How intrepid are journal editors required to be?

RELMAN: You mean, should editors raise the question with the institution when they have some suspicion of fraud? Yes, they should. But it doesn’t happen very often, Must the Lab chief Always Be an Author?

Q: Should the head of lab or research team always get authorship credit? RELMAN: Absolutely not. Authorship should imply some meaningful intellec- tual contribution to the work—not simply providing the resources, the laboratory or the money.

Q: Is that the Journal’s policy?

RELMAN: Oh, yes.

Q: How is that enforceable? As you said, you are not a policeman.

RELMAN: We don’t enforce it, we simply espouse it. I have written on what authorship means. We endorse the uniform requirements document of the International Committee of Medical Journal Editors, which makes a very strong statement on that issue. When we see an article with many names on it, we always question the authors about whether they all made significant contributions. But if we get a paper with four names, we have no way of knowing that only one did the work and the other three are getting a free ride. Some journals have a policy of asking the authors to sign a statement saying that they not only all approve of the submission of the article, but that they all contributed to it. That’s not a bad idea.

Q: You don’t at present have such a policy?

RELMAN: We don’t—not because we oppose it, but because we simply haven’t gotten around to thinking about it. We’ve been accused of being so controlling and dominating that I hesitate to make yet another demand on our authors. So we’ve had to pick and choose where we will take our stand and we haven’t yet chosen to do that. But we may, because I think it’s a legitimate consideration and it’s much abused. It’s part of the fraud problem. People put their names on papers when they haven’t taken any responsibility for supervising the work. They’ve allowed some junior person to do it and reward them with authorship, whereas they should be in the laboratory watching what’s going on.

Relman, who received his M.D. degree from Columbia University in 1946, has been active in medical practice and education as well as in publishing. He was a professor at Boston University and the University of Pennsylvania, where he chaired the department of medicine. He returned to Boston in 1977 to become editor of the Journal; he also is a senior physician at Brigham and Women’s Hospital and a professor of medicine at Harvard Medical School.

He was interviewed on February 25 by Tabitha M Powledge, editor of THE SCIENTIST, at his Journal office in Boston. This is an edited version of their talk.

Q: You have been much in the news recently and criticized rather a lot. How do you feel about that?

RELMAN: That’s nothing new. The Journal has been prominent for a long time. We're often involved in controversial issues of broad public interest. I have been known to take controversial stands on some of those issues, so I’ve come in for my share of criticism. But I must admit the last few weeks have been rather more hectic than usual.

Q: Is that the appropriate role for a journal editor, to set standards?

RELMAN: It seems to me that the Journal—by its very nature and by its position in American medicine—is bound to be controversial. We are the oldest and, certainly on this continent, the best-known medical journal. We’ve always carried articles that were not only of great scientific interest, often controversial, but articles on social and political and economic and ethical issues.

We consider ourselves to be an open forum for the discussion of issues that are important to medicine and health. As the moderator of that forum, I have to make choices about what voices are going to be heard, and that inevitably is going to draw criticism. However, I have from time to time spoken out on issues of the day, such as handguns in American society, the medical industrial complex and the commercialization of health care, ethical issues affecting doctors, and the intrusion of the courts into the practice of medicine. I speak as an individual. I make it clear that those opinions are not the official views of the Journal. Nevertheless, I’m afraid these views attract attention.

Some people have suggested that perhaps I ought not to speak out on some of these issues, that as editor I ought to try and be neutral. But it seems to me that would be contrary to what these times demand. People who have convictions, and who think they can bring some reasonable discussion to the debate, have an obligation to speak out. What I must be very careful about, of course, is to keep the Journal always open to all points of view, and that’s what I scrupulously try to do. I don’t allow my personal views to influence my choice of material for the Journal.

Q: The New England Journal’s importance is indisputable. But is the Journal too important, too powerful?

RELMAN: I don’t know how to respond to that question. Any influence the Journal has must be the result of the trust and the confidence reposed in it by the leaders of American academic medicine—by investigators, by teachers and by physicians. We actively seek no controlling role. We don’t try to influence things other than to tell people what our standards are. If they wish to support them, that’s fine. And if they don’t, the world is full of journals. We certainly have plenty of good competition.

Most of the criticism of our policies comes not from our readers or from our contributors but from the media. The Journal is published not for the media but for the medical profession. If what we write is newsworthy, fine. But inevitably the media’s interests and ours are not always consonant.

Newsmen look at us simply as a source of news, and they sometimes get impatient with policies that we know our readers want us to follow, and that our contributors support. If our readers didn’t want us to stay with those policies, we would change them. We are serving the medical profession, and through them we serve the public.

Q: You speak of policies. Are you referring specifically to the embargo?

RELMAN: And the Ingelfinger rule, yes. Those are two separate but related policies. The embargo is an arrangement we make with the media. The Ingelfinger rule is an arrangement we make with our contributors. Neither will work unless both parties to the agreement are willing to go along with it. If the press will not accept the embargo, it won’t work. And if our contributors don’t want to abide by the Ingelfinger rule, we’ll be forced to abandon it. But our reading is that the vast majority of our readers and contributors support these policies. yet. They have said they intend to, and they may have done so.

Q: There’s been a lot of speculation in the press that other publications will also. Do you know of such cases?

RELMAN: No, I don’t I’m hoping very much that this will not occur. I’m hoping that Reuters will not want to do this very often, that they will find it possible to do it even less often than that, and that when they do it, the rest of the media will stand firm. I took this action against Reuters most reluctantly, and only after talking to them at length. I then wrote a letter to all of the media subscribers to the Journal explaining what I had done, reviewing once again the reason for the embargo and expressing my hope that they would continue to stay with it. I’ve had some positive responses. News people have contacted me to say they intend to observe it, and they thought that what I had done is right.

Q: How do you feel about that decision now?

RELMAN: In principle what I did was right. One could argue that as a tactical move maybe it wasn’t the most adroit thing I could have done, but they have been so adamant. I have had second thoughts.

I felt impelled to do it because it seemed to me that, despite their protests, it was a flagrant violation of the embargo. They said they got the information from other sources. But the fact is they had the Journal on their desks at the time they made the decision to go with the story. They knew it was coming out in the Journal; in fact the story they put out on their wires was to the effect that this was coming out in the Journal. Furthermore, the information they acted on was widespread It was on the business wires aleady. Many other news people had the same information, and they chose not to go with it because they wanted to observe the embargo. Only Reuters broke it.

When they did break it, I received a lot of adverse comment from other people in the media and from my colleagues here at the Journal saying, in effect, “Do you have an embargo policy or don’t you? If you have a policy you have to enforce it. If you don’t enforce it, it will be meaningless.” So with those kinds of pressures, and in view of the enormous publicity attending the aspirin story, I decided I had to do something. Perhaps I should have tried to reach some compromise initially. But in any event, having heard their story, and, having decided that they clearly had broken the embargo, I did what I said would do: I withdrew their first-class subscription.

When I went back to them after that with the idea that maybe we could reach some compromise, I didn’t want a confrontation. I want to cooperate with them as much as we can. I said, “If I reinstate you, would you agree not to break our embargo in the future?” They said no. Q: Are negotiations continuing?

RELMAN: Not at the moment. I hope, when things cool down a bit, that we can talk again. There is no reason why we should be at war over a thing like this.

There is this about Reuters, though, that may make a compromise more difficult. Unlike most of the other media agencies that get the Journal, Reuters is strongly oriented to the business community They have come to feel, evidently, that their primary obligation is to their clients who are investors and stock analysts and businessmen. Their executive editor has told me that any information that influences or that might influence the stock market is fair game, and they will publish that information regardless of the source, regardless of an embargo.

What’s not yet clear is whether they intend to follow that policy only with respect to New England Journal articles or whether they are going to do this generally. If they are going to apply that to articles from the Journal only, then it seems obvious that what they’re doing is conducting a war with us. It’s not a matter of principle; they simply are retaliating for our withdrawal of their subscription.

If they apply this policy to all medical journals, then we are in a new era. What we’re then seeing is the commercialization or privatization of scientific news. The usual standards that have been applied to the way scientific news is disseminated are going to be set aside in the interests of the market because science is now so important to the market. That would be very unfortunate.


Q: It appears, though, that there are a number of recent cases of information being available before publication. For example, stock analysts and a few investors have been able to get and profit from information be forehand. I assume you know of specific cases like that, not just at the New England Journal but at other journals as well?

RELMAN: Oh yes. Unfortunately, too many. Q: What can you do to control that? Is it even appropriate to try?

RELMAN: Those are difficult questions. We do not send advance copies to stock analysts or brokerage firms, and therefore we have no embargo agreement with them as we have with the media. So in one sense we have no right to expect the stock analysts firms and other business sources who manage to find out what’s coming out in the Journal in advance to observe the embargo. Having said that, I must admit that I find it very distressing. We would try to stop it if we could legally and ethically do so, and if it were practical. We’re trying.

One source of leaks, of course, is simply the early receipt of regular copies of the Journal. The Journal comes off the press a week in advance of publication and is put in the mail to our regular subscribers. Depending on where you live and the vagaries of the U.S. mail delivery, you may get your Journal delivered as much as four or five days in advance of the publication date. It’s easy to imagine that someone who is really interested can get a copy early. We’re looking into that to see what can be done.

Q: You think the stock-related leaks are coming from early delivery of regular copies, not being leaked by journalists who, have gotten early copies?

RELMAN: I have no way of knowing for sure, but I’m inclined to believe that the vast majority of reporters are quite conscientious about the embargo. They want to observe it. It’s to their advantage. The leaks are probably not coming from that source.

One obvious possibility is the people connected with the study. If a company supplied the product that’s being studied or is making the drug or whatever, the company may hear about the study before it’s published. If we can legally and ethically prevent those kinds of leaks, we should. We are going to try to do what we can by changing the mail delivery, by making sure that press conferences don’t admit stock analysts, and by reviewing even more carefully with authors and sponsors of research in press what their obligations are.

But we have to recognize that if someone is determined to dig out this information, because so many people are involved in the production of the information, leaks are going to occur. We’re not the CIA, we’re not the FBI, and we have no authority nor any desire to act as policemen. We’ll do what we can because it is in the public interest to have some sort of uniform, orderly system for release of information. Anything else damages doctors’ abilities to take care of patients properly. It’s also bad for patients because they can’t find out the answers to their questions promptly.

So a uniform release time, the embargo system, is right for the medical profession. It’s right for patients. It makes sense for the news media. It gives them plenty of time to research their stories in depth and it avoids the chaos of cutthroat competition.

Q: What have been the consequences of your decision not to not send airmail copies to Reuters? Have they broken embargoes as they threatened?

RELMAN: I cannot say for sure. I am told that they have, but I haven’t seen the evidence and it’s a very difficult call. Take the case of John Darsee. We published two of his papers that turned out to be in part fraudulent. One of those papers had some data involving renal function studies on patients. In my former life, I was a renal physiologist and nephrologist. I’m quite familiar with the methods that were used and could be considered an expert referee in that area. I said to myself at the time, what beautiful data he has. Those results are cleaner and nicer than anything I achieved in 30 years in the field. It crossed my mind, inevitably, could he have made up those numbers? But I dismissed those thoughts. What could I do? Should I have written to him and said, are you really telling me the truth?

Q: But that was then. Presumably everybody is a little more cynical and suspicious now than they were then. Would your behavior be different now, too?

RELMAN:Yes. I’m a little more sensitive. It would take less to make me suspicious and ask more questions than I would have five or 10 years ago. But what I want to emphasize is that still that’s not going to guarantee the detection of fraud at the journal level. The journal is not the place to detect fraud. Fraud should be detected where the work is being done, by co-workers and supervisors. Otherwise, the only other time it’s likely to be discovered is after it’s published, when people try to replicate the work or follow implications that don’t work out. Q: Do you concur with your senior deputy editor, Marcia Angell, that if a limit was set on the number of papers scientists could list on a c.v. when applying for funding or a promotion, it would be an advance toward quality control?

RELMAN: Yes, I do. It wouldn’t be the solution entirely, but I think it would help. There is no question that the pressure to pad your bibliography, the obsession with the number of papers, plays an important role in driving people to do sloppy work. It also drives them to do trivial things and to publish repetitive and overlapping papers.

Q: Is the tremendous growth in the number of journals a factor?

RELMAN: Yes. Of course, it’s a loop. The proliferation of journals reflects the proliferation of science. Subdivisions of science are created, and the output from that needs to find new journals. On the other hand, I know of many instances when the founding of a new journal was not in response to a manifest need but to a corporate need. A professional society wants to have a journal for one reason or another. Or, more often, a publisher wants to have more journals. You know as well as I do the tricks of the publishing game. You can make money by publishing a journal that isn’t very successful providing you can get a certain minimum captive circulation and some advertising. Scientific need is not always the dominant consideration. Often it is simply a cold-blooded financial calculation.

Q: Can anything be done? Or does one just have to allow journals to proliferate?

RELMAN: I’m afraid so. In our free society I don’t see how you can stop it. What one can do is to point out the problem and try to maintain some standards. Most responsible editors try to oppose needless proliferation of the literature, needless duplication and splitting, and production of articles simply for the sake of production.