J T King1. 1. Center for Health Services Research, VA Pittsburgh Healthcare System, and Department of Neurological Surgery, University of Pittsburgh, Pennsylvania, USA. jking@neuronet.pitt.edu
Abstract
OBJECTIVE: Authorship proliferation in biomedical research has become rampant; the proportion of single-author articles is decreasing, the percentage of multiauthor articles is increasing, and the number of authors per publication is increasing. To determine whether authorship trends in the North American neurosurgical literature parallel trends observed in other areas of the biomedical literature, I studied original neurosurgical research articles published in the past 50 years. METHODS: I sampled clinical, anatomic, and laboratory investigations in Neurosurgery and the Journal of Neurosurgery at 10-year intervals from 1945 to 1995. For each research article, I determined the number of authors, the countries of origin, and the educational degrees of the authors. RESULTS: The mean number of authors per article has increased steadily in the past 50 years, from 1.8 (standard deviation, +/-0.8) authors per article in 1945 to 4.6 (standard deviation, +/-2.1) authors in 1995. The proportion of single-author articles is decreasing; these articles accounted for 43% of articles in 1945 and only 3% of articles in 1995. Increases in the proportions of non-M.D. authors and of articles originating outside the United States were also observed. CONCLUSION: The proliferation of multiauthor articles and the decrease in the proportion of single-author articles in the neurosurgical literature parallel trends observed in many other areas of biomedical research. Possible explanations include larger research teams, variable or absent journal criteria for defining authorship, and the increased awarding of "gift" authorships.
OBJECTIVE: Authorship proliferation in biomedical research has become rampant; the proportion of single-author articles is decreasing, the percentage of multiauthor articles is increasing, and the number of authors per publication is increasing. To determine whether authorship trends in the North American neurosurgical literature parallel trends observed in other areas of the biomedical literature, I studied original neurosurgical research articles published in the past 50 years. METHODS: I sampled clinical, anatomic, and laboratory investigations in Neurosurgery and the Journal of Neurosurgery at 10-year intervals from 1945 to 1995. For each research article, I determined the number of authors, the countries of origin, and the educational degrees of the authors. RESULTS: The mean number of authors per article has increased steadily in the past 50 years, from 1.8 (standard deviation, +/-0.8) authors per article in 1945 to 4.6 (standard deviation, +/-2.1) authors in 1995. The proportion of single-author articles is decreasing; these articles accounted for 43% of articles in 1945 and only 3% of articles in 1995. Increases in the proportions of non-M.D. authors and of articles originating outside the United States were also observed. CONCLUSION: The proliferation of multiauthor articles and the decrease in the proportion of single-author articles in the neurosurgical literature parallel trends observed in many other areas of biomedical research. Possible explanations include larger research teams, variable or absent journal criteria for defining authorship, and the increased awarding of "gift" authorships.