Editors have a duty to act if they suspect misconduct by an author.1 However, since they are not equipped to investigate misconduct allegations or impose punishment they must rely on the author’s institution.2 The Committee on Publication Ethics’ (COPE) code of conduct states that “Editors should make all reasonable efforts to ensure that a proper investigation is conducted.” However, the code also hints at the difficulties editors may face, stating: “if this does not happen, editors should make all reasonable attempts to persist in obtaining a resolution to the problem. This is an onerous but important duty.”1 I have used the COPE database of cases (box) to show the difficulties editors may face when they try to alert institutions to possible cases of research or publication misconduct.

Cases presented to COPE cannot be used to estimate the frequency of problems. Members probably bring only their most troublesome cases to the COPE forum, and so the database is likely to over-represent cases where the response from institutions has been unsatisfactory. Furthermore, COPE hears only the editors’ side of the story.