Bibliometric analysis of manuscript characteristics that influence citations: A comparison of four major neurology journals

Objective: To inspect 28 data characteristics among the top neurology journals with the highest impact factor and their influence on citation rate. Methods: Consecutive articles from January 2004 to June 2004 were collected from four major neurology journals with the highest impact factor: The Lancet Neurology (impact factor, 11.964), Acta Neuropathologica (7.589), Brain (5.858) and Annals of Neurology (5.706). Web of Science was used to extract the citation count for these articles, and 28 article characteristics were tabulated manually. Univariate analysis and a multiple regression model were performed to predict citation number from the collected variables. Results: A total of 288 manuscripts i.e. 24 in The Lancet Neurology, 70 in Acta Neuropathologica, 117 in Brain and 77 in Annals of Neurology. Univariate analysis revealed the following variables to have a significant positive correlation with increased citations: journal (1; p<0.0001), country of origin (15; p<0.0001), number of tables (28; p=0.0007), words per title (7; p=0.0006), design of study (17; p=0.001), open access (22; p<0.0001), total words (24; p<0.0001), total references (25; p<0.0001) and total number of pages (26; p<0.0001). In a multivariate regression model the following variables predicted increased citation count (p < 0.0001, R2 = 0.4377): number of pages, open access status, multicenter studies and journal origin. Conclusion: The results of our bibliometric study may be used by authors while compiling their manuscript to increase recognition and improve the impact of their articles over what they would normally experience.