Correspondence
. . . And the Journal's Policies
N Engl J Med 1994; 330:503-504February 17, 1994
- Article
To the Editor:
I take issue with the statement by Kassirer and Angell (Aug. 19 issue)1 that “financial conflicts of interest are a matter of choice.” This implies that biomedical research worthy of publication takes place only in academic institutions and ignores the fact that important research is not only financed, but also performed, by commercial companies.
Stringent policies on the authorship of review articles disenfranchise scientists in industry and may paradoxically lead to increased secrecy, because they reinforce the view that publications by such employees lack credibility. Such policies may also exclude the opinions of acknowledged experts: Paul Ehrlich could not have written about salvarsan since he was sponsored by Hoechst, and Sir James Black, who developed both beta-blockers and H2 antagonists, would also be silenced.
Conflict of interest is most dangerous when it is not acknowledged or when it is even hidden. Journals are therefore best served not by blanket rules that disqualify certain authors but by the peer-review system in which each article, whether it is a literature review or original research, is judged on its merits. It is important for referees, readers, and editors to have as much information as possible, including acknowledgment of who actually did the work, so that they may draw their own conclusions about bias, financial or intellectual, unintentional or not.
1 ReferencesElizabeth Healing
Janssen Pharmaceutical, Grove OX12 0DQ, United Kingdom1
Kassirer JP ,Angell M . Financial conflicts of interest in biomedical research. N Engl J Med 1993;329:570-571
Full Text | Web of Science | Medline
To the Editor:
The implementation of policies on conflict of interest in a scientific journal has ramifications that are not acknowledged in the article by Thompson1 or the accompanying editorial by Kassirer and Angell2. I offer here two criticisms of current policies at the Journal.
The editors request information from all their contributors about potential conflicts of interest on submission of a manuscript. Kassirer and Angell2 assure us that this information is not used in the editorial-review process; it is purportedly used only to determine what disclosures will be needed if the paper is published. Yet the Journal's policy requires that editors have the information in front of them during the editorial review. If the editors are sincere in their claim that the information they request is not used to assess submissions, they should request it from authors after a decision about publication has been made, when it could not influence the decision. As with reprint orders, information on potential conflicts of interest could be solicited after the acceptance of a paper for publication.
Kassirer and Angell also assert that the Journal's policy of refusing to publish editorials and review articles by certain people is not censorship. They claim that this policy is no different from the editors' prerogative to choose what gets published: just as studies with animals as subjects might be rejected as unsuitable for the Journal, so would articles by people whom the editors consider to have a financial conflict of interest2.
The problem with this argument is that there is a difference between rejecting a submission because of its content and rejecting a submission without looking at the content, simply on the basis of who wrote it. This policy is tantamount to creating a blacklist for each review or editorial topic. Even if the policy does occasionally protect readers from reading the work of biased authors, like all blacklists it carries the burden of suppression of opinion that must be weighed against the purported benefits3. Where is the evidence that the benefits of protecting readers from dangerously slanted writings -- the excuse for censorship whenever it has been imposed -- outweigh the harm caused by the blacklist? I suggest that in the absence of hard evidence, the readership of the Journal deserves a less rigid policy from its editorial staff.
3 ReferencesKenneth J. Rothman, Dr.P.H.
Epidemiology, Newton Lower Falls, MA 02162-14501
Thompson DF . Understanding financial conflicts of interest. N Engl J Med 1993;329:573-576
Full Text | Web of Science | Medline2
Kassirer JP ,Angell M . Financial conflicts of interest in biomedical research. N Engl J Med 1993;329:570-571
Full Text | Web of Science | Medline3
Rothman KJ . Conflict of interest: the new McCarthyism in science. JAMA 1993;269:2782-2784
CrossRef | Web of Science | Medline
Author/Editor Response
The editors reply:
Dr. Rothman makes too much of it. We do not see ourselves as protecting readers from “dangerously slanted writings,” and stirring words like “censorship” and “blacklist” are misplaced in this context. We have simply decided that review articles are best written by people without financial conflicts of interest, because they are more likely to be dispassionate in evaluating the literature. This is not censorship, any more than it is censorship to forbid a judge to decide a case involving a company in which the judge has a financial interest.
The policy for scientific reports is different, in part because the presence of original data makes it easier to discern bias and deal with it. We do not inform peer reviewers of authors' financial conflicts of interest. Although, as Dr. Rothman points out, the information is available to the editors, we hardly require that they “have it in front of them during the editorial review.” Indeed, in the great majority of cases, this information plays no part, much less an overriding one, in our deliberations.
Ms. Healing is correct that commercial companies perform useful research. That is a different matter from the question of whether researchers for these companies are the appropriate authors of review articles on related subjects.
Jerome P. Kassirer, M.D.
Marcia Angell, M.D.
