Secondary Logo

Journal Logo

Is There a Dark Phase of This STROBE?

MacMahon, Brian*; Weiss, Noel S.†‡

doi: 10.1097/EDE.0b013e3181570f6e

Suggestions for improving the reporting of observational epidemiologic studies are to be encouraged, but we believe the required use in reports of a checklist of design and analytic features relevant to study quality would have detrimental consequences.

From the *Department of Epidemiology, Harvard School of Public Health, Boston, Massachusetts; †Department of Epidemiology, University of Washington; Seattle, Washington; ‡Fred Hutchinson Cancer Research Center; Seattle, Washington.

Editors' note:Related articles appear on pages 789, 792, 794, 797, 800, and 805.

Correspondence: Brian MacMahon, Department of Epidemiology, Harvard School of Public Health, Boston, MA. E-mail:

Proposals to improve the reporting of epidemiologic and other scientific studies are always welcome, but we have reservations about the usefulness of the document developed by von Elm and colleagues.1

One of the purposes of the document is said to be “to support editors and reviewers when considering such articles for publication.” This smacks of condescension. Typically, the editors and reviewers chosen by a journal are not newcomers: it is our impression that the editors of epidemiologically-oriented journals are well informed about the issues referred to in the von Elm document. Editors of other major medical journals, if they are not themselves epidemiologists, have consultants who are. Such people are not usually in need of a checklist to perform their task.

What would be the effect on prospective authors of wide-scale use of a check-list for submitted manuscripts? For the novice, it might serve a purpose, but there are potential downsides. Certainly, the principles that underlie such a list are matters that should be in the forefront of the design, conduct, and analysis of all studies—surely not to be introduced at the time of reporting. Is there a risk that an author who becomes aware of these principles only at the stage of report writing will shade his or her description towards what should have been done rather than what was done? While appreciating the value of clear and accurate prose, we (like most readers) try to avoid judging an apple by how well it is polished.

Potential reviewers, aware of the time and effort required to determine whether the work conforms to a lengthy list of specified characteristics (33 items in 22 categories) may be more inclined to decline invitations to review. Moreover, having accepted an assignment, the reviewers' efforts to deal with the checklist may come at the cost of time and effort spent evaluating more broadly the real value of the work.

We concur with von Elm and colleagues that the reporting of epidemiologic studies should be transparent “so that readers can follow what was planned, what was done, what was found, and what conclusions were drawn.” We concur also that reporting sometimes falls short of this standard. Nonetheless, we believe that the prescription that these authors have written for improving the present state of affairs could lead to adverse effects that are as or more serious than the problem that prompted the prescription.

Back to Top | Article Outline


BRIAN MACMAHON is professor emeritus in the Department of Epidemiology of the Harvard School of Public Health, where he has been since 1958. NOEL S. WEISS has been an epidemiologist at the University of Washington and the Fred Hutchinson Cancer Research Center since 1973.

Back to Top | Article Outline


1. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. Epidemiology. 2007;18:800–805.
© 2007 Lippincott Williams & Wilkins, Inc.