Coauthorship in pathology, a comparison with physics and a survery-generated and member-preferred authorship guideline.

In a large and detailed survey of scientific coauthorship in pathology, 3500 members of the US and Canadian Academy of Pathology (USCAP) were surveyed via the Internet with a final response rate of 22.5%. The results were compared with a previous survey of members of the American Physical Society (APS). The fields are found to be very similar. For example, there is no well-defined way to determine coauthorship: the byline is arrived at without the use of public coauthorship standards according to 90% of respondents (92% in physics). A substantial amount of inappropriate authorship is present in both fields using a variety of authorship guidelines. For example, using the guideline of the International Committee of Medical Journal Editors (the "Uniform Requirements for Manuscripts Submitted to Biomedical Journals" [ICMJE]), the average number of coauthors judged to be inappropriate in pathology on papers with 4 coauthors is 1.0 (1.2 in physics), and using the guideline requiring "direct contributions to scientific discovery or invention," we find 1.6 (1.5 in physics). Finally, it is suggested that authorship guidelines should be constructed by public surveys rather than closed-door committees: an authorship guideline constructed from previous survey feedback (from APS members) was found to be preferable to USCAP members (it received 40% of the vote, the ICMJE received 24% of the vote).