RATING AUTHORS CONTRIBUTIONS TO COLLABORATIVE RESEARCH - THE PICNIC SURVEY OF UNIVERSITY DEPARTMENTS OF PEDIATRICS

Objectives: To determine how department chairs in pediatrics rate invo lvement in medical research and to determine whether faculty deans' of fices have written criteria for evaluating research activity when asse ssing candidates for promotion or tenure. Design: Cross-sectional mail ed survey and telephone survey. Setting: Canadian faculties of Medicin e. Participants: Chairs of the 16 Canadian university departments of p ediatrics and deans' offices of the 16 university medical faculties. M ain outcome measure: Weight assigned by department chairs to contribut ions to published research according to author's research role and pos ition in list of authors and the method of listing authors. Results: F ifteen of 16 chairs responded. Twelve submitted a completed survey, tw o described their institutions' policies and one responded that the in stitution had no policy. Eleven reported that faculty members were per mitted or requested to indicate research roles on curricula vitae. The re was a consensus that all or principal investigators should be liste d as authors and that citing the research group as collective author w as insufficient. The contribution of first authors was rated highest f or articles in which all or principal investigators were listed. The c ontribution of joint-principal investigators listed as first author wa s also given a high rating. In the case of collective authorship, the greatest contribution was credited to the principal investigator of th e group. Participation of primary investigators in multicentre researc h was rated as having higher value than participation in single-centre research by seven respondents and as having equal value by four. Only one dean's office had explicit written criteria for evaluating author ship. Conclusions: Most departments of pediatrics and medical faculty deans' offices in Canadian universities have no criteria for assessing the type of contribution made to published research. In view of the t rend to use multicentre settings for clinical trials, guidelines for w eighting investigators' contributions are needed.