Correspondence

Disclosure of Authors' Conflicts of Interest — A Follow-up

N Engl J Med 2000; 343:146-147July 13, 2000

Article

To the Editor:

In the February 24th issue, 19 cases were identified in which authors of Drug Therapy articles were in violation of the Journal 's conflict-of-interest policy for review articles and editorials.1 According to this policy, “authors of such articles will not have any financial interest in a company (or its competitor) that makes a product discussed in the article.”2 Although no wrongdoing was implied, one cannot look at this matter without thinking that, in some small way, the authors have been discredited.

There are several problems with the Journal 's analysis of these cases. First, only the Drug Therapy articles have been reviewed for violation of the conflict-of-interest policy. However, the policy applies equally to other types of review articles and editorials. Why was the policy not enforced for these articles as well? The Journal has focused on pharmaceutical companies, but clearly, review articles and editorials related to biotechnology, diagnostic tests, medical devices, new procedures, and so forth should be examined in the same light.

Kelly M. McMasters, M.D., Ph.D.
University of Louisville Brown Cancer Center, Louisville, KY 40202

2 References

1
Angell M, Utiger RD, Wood AJJ. Disclosure of authors' conflicts of interest: a follow-up. N Engl J Med 2000;342:586-587
Full Text | Web of Science | Medline

2
Information for authorsN Engl J Med 2000;342:585-585

To the Editor:

The Journal 's policy on financial conflicts of interest seems too restrictive. The authors most qualified to write Drug Therapy articles are usually the same ones chosen to participate in the development of drugs. Thus, it would seem that the Journal 's policy reduces the quality of reviews by disqualifying the best authors. Why should the editors assume that persons involved in the development of drugs are automatically too biased to write Drug Therapy articles, particularly if the authors divulge their potential conflicts of interest? Readers can be trusted to decide whether there is bias. Furthermore, eliminating any author who has ties to companies that make drugs does not guarantee the elimination of bias.

Robert R. Recker, M.D.
Creighton University School of Medicine, Omaha, NE 68131

To the Editor:

The Journal has admitted that 19 Drug Therapy articles published since January 1997 were in violation of its published conflict-of-interest policy and that the authors should have been disqualified. Considering that the authors of these articles were financially involved with relevant companies, the Journal should retract these articles.

Billi Goldberg, M.P.A.
International Dinitrochlorobenzene Group, San Francisco, CA 94131

Author/Editor Response

The editors reply:

The views expressed by McMasters, Recker, and Goldberg are representative of the divergent opinions about whether it is acceptable for authors of review articles and editorials to have financial relationships with for-profit companies that make products discussed in the articles. We think they should not have such relationships, as summarized in a 1996 editorial.1 We did not mean to discredit the authors listed in our letter in the February 24th issue. We had erred in applying our policy to Drug Therapy articles, and we thought we should set the record straight. It goes without saying that the policy applies to relationships with all types of companies, not simply pharmaceutical companies.

We do not think, as Recker suggests we do, that authors who have such relationships are automatically biased, but why should readers even have to wonder? Furthermore, we think highly qualified authors who have no financial relationships with for-profit companies can almost always be identified.

Goldberg recommends that the articles be retracted, but that would not be appropriate in this case. Retraction is appropriate only when work has been proved to have been falsified or plagiarized or, in the case of Original Articles, when an error has been discovered that renders the work invalid.

Marcia Angell, M.D.
Robert D. Utiger, M.D.
Alastair J.J. Wood, M.D.

1 References