Authorship: to be or not to be?

Abstract

Authorship policies at medical journals vary substantially and journals can be ambiguous as to what is expected of authors. International Committee of Medical Journal Editors (ICMJE) criteria emphasize two components of authorship—credit and responsibility—in a complex multipart definition. Yet ICMJE criteria are problematic because they also require final manuscript approval. If final approval is ceded to an academic “key opinion leader,” it can be rationalized that this requires naming that person as author, when they may have had little role in content creation. We propose a new criterion; an author is someone who has free and unfettered access to all raw data. Such access is essential to independently test or verify hypotheses, but it would also potentially permit data to be manipulated. Unfettered access to raw data is not sufficient for authorship but it is necessary, since the validity of data can be determined in no other way