Objectives To estimate the prevalences of perceived honorary authorship and International Committee of Medical Journal Editors (ICMJE)-defined honorary authorship, and identify factors affecting each rate in the physical medicine and rehabilitation literature. Design Internet-based survey. Setting Not applicable. Participants First authors of articles published in 3 major physical medicine and rehabilitation journals between January 2009 and December 2011 were surveyed in June and July 2012 (N=1182). Interventions Not applicable. Main Outcome Measures The reported prevalences of perceived and ICMJE-defined honorary authorship were the primary outcome measures, and multiple factors were analyzed to determine whether they were associated with these measures. Results The response rate was 27.3% (248/908). The prevalences of perceived and ICMJE-defined honorary authorship were 18.0% (44/244) and 55.2% (137/248), respectively. Factors associated with perceived honorary authorship in the multivariate analysis included the suggestion that an honorary author should be included (P<.0001), being a medical resident or fellow (P=.0019), listing "reviewed manuscript" as 1 of the nonauthorship tasks (P=.0013), and the most senior author deciding the authorship order (P=.0469). Living outside North America was independently associated with ICMJE-defined honorary authorship (P=.0079) in the multivariate analysis. In the univariate analysis, indicating that the most senior author decided authorship order was significantly associated with ICMJE-defined honorary authorship (P=<.001). Conclusions Our results suggest that honorary authorship does occur in a significant proportion of the physical medicine and rehabilitation literature. Additionally, we found several factors associated with perceived and ICMJE-defined honorary authorship and a discrepancy between the 2 rates. Further studies with larger response rates are recommended to further explore this topic.
- Medical ethics