Objective: Many biomedical journals have introduced disclosure of research contribution in journal articles as a way to limit irresponsible authorship. Over the years, journals have developed different contribution disclosure policies and procedures, but it is not known how these policies and procedures affect authorship. The aim of our study was to analyse possible causative relationships between the structure of the contribution disclosure form and the number of authors who do not meet authorship criteria in biomedicine set by the International Committee of Medical Journal Editors (ICMJE).
Research design and methods: In a single-blind randomized trial, 1462 authors of 332 manuscripts submitted to a general medical journal answered one of three different contribution disclosure forms: (1) open-ended -- asking the respondents to describe in their own words their contributions in the submitted work; (2) categorical -- with 11 possible contribution choices and (3) instructional -- guiding through the ICMJE definition and instructing how many contributions are needed to satisfy individual criteria.
Main outcome measure: The number of authors not satisfying ICMJE authorship criteria.
Results: The group answering the instructional form had significantly fewer authors whose reported contributions did not satisfy ICMJE criteria (18.7%) than the groups answering the categorical (62.8%) or open-ended (54.7%) form (chi2(2 ) = 210.833, p < 0.001). The group receiving instructional forms also had the fewest articles with honorary authors (32.5%) compared with those who responded in a categorical (68.7%) or open-ended (83.0%) form (chi2(2 ) = 63.378, p < 0.001). All authors who answered the open-ended form, regardless of their compliance with authorship criteria, reported significantly fewer contributions (median 3 [95% confidence interval = 3-3]) than authors responding to either the categorical (4 [4-4]; z score = -7.1899, p < 0.001) or instructional form (4 [4-5]; z score = -13.9760, p < 0.001). Honorary authors answering the instructional form reported more contributions than those answering either the categorical or open-ended form (z score = 2.7637, p = 0.0057 and z score = -3.3773, p < 0.001, respectively). Most honorary authors (39.9%) lacked only the third ICMJE criterion (final approval of the submitted manuscript).
Conclusions: The structure of the contribution disclosure form significantly influenced the number of contributions reported by authors of submitted manuscripts and their compliance with the ICMJE authorship criteria. Journal editors should be aware of the cognitive aspects of survey methodology when they construct for their authors self-reports about behaviour, such as contribution disclosure forms.