Authorship changes when publishing the works presented in national student forums, Cambios de autoría al publicar los trabajos presentados en fórums nacionales estudiantiles
Description
Introduction: The problems with the articles' authorship are frequent; however, in Cuba there are few investigations on the topic. Objective: To determine the authorship changes in the publication of works presented in medical student forums at national level in Cuba and its associated factors. Methods: Observational and cross-sectional study using analytical techniques, that included all the works presented in 2016, 2017 and 2019 editions of the medical student forums at national level. A Google Scholar search strategy was carried out in order to check if these works had been published in scientific journals. After, authorship changes were identified and associated factors were looked for. Prevalence ratios (aPR), 95 % confidence intervals (95 % CI) and p values using generalized linear models were obtained. Results: Of the 129 published works, 89,9 % (n= 116) had authorship changes. In the multivariate analysis, there was more frequency of authorship changes when the articles were published in professionals' journals (aPR: 1,29; IC 95 %: 1,11 - 1,50; p= 0,001) and when the publication occurred later to the event (aPR: 1,20; IC 95 %: 1,01 - 1,42; p= 0,042); that which was contrary when the corresponding author was a student (aPR: 0,79; IC 95 %: 0,66 - 0,95; p= 0,010), adjusted by three variables. Conclusion: Almost the entirety of the articles had authorship changes and some factors associated to a higher or smaller frequency in those changes were determined (publication, time and corresponding author).
Files
Steps to reproduce
An observational and cross-sectional research was conducted using analytical techniques, which included all the works presented in the 2016, 2017, and 2019 editions of the FNCM. In the first two editions, the in-person sessions took place at the medical universities of Pinar del Río and Cienfuegos, respectively, while in 2019, the event was held exclusively in a virtual format. Full-text papers (in PDF format) were collected, and their availability was verified on the event promotion website (http://promociondeeventos.sld.cu) belonging to Cuba's health telematic network (Infomed http://www.sld.cu). To identify works published in scientific journals, a search strategy was employed in both Spanish and English through Google Scholar (http://scholar.google.com.cu). This strategy involved a combination of terms, including the study's topic and location, as well as the first, second, and last author's names, in the following format: ("study topic" AND "study location") AND (author: first OR author: second OR author: last). This strategy was chosen for its demonstrated usefulness in similar research(6,7,8), given Google Scholar's (GS) ability to identify articles and journal citations. The searches were conducted between June 10 and June 15, 2021. To declare that a work was published, there had to be a match between the presented work and the published article in terms of the topic, study location, and main results. 17 works that appeared in abstract books of other student events were not considered as published, as they were published in Cuban student journals and only included title, authors, and abstract. A database was generated using Microsoft Excel (version 2013 for Windows). Descriptive analysis of categorical variables (event edition, variation in the number of authors, type of journal, articles by country, typologies, design, corresponding author, language, and year of publication) was conducted using absolute and relative frequencies. The quantitative variable (number of authors) was described with the best measure of central tendency and dispersion. For the preparation of Table 2, Student's t-tests (for the variable number of authors) and Fisher's exact tests (for other variables) were used. Crude and adjusted prevalence ratios (RPc and RPa), 95% confidence intervals (CI 95%), and p-values were obtained using generalized linear models (Poisson family, log-link function, and models for robust variances). Values with p≤ 0.05 were considered statistically significant. To include a variable in the multivariate model, p-values had to be ≤ 0.30 in bivariate models. Stata v.11.1 statistical software (StataCorp LP, College Station, TX, USA) was used.