RT Journal Article
SR Electronic
T1 Evaluation of Reproducibility in Urology Publications
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 773945
DO 10.1101/773945
A1 Rauh, Shelby Lynn
A1 Johnson, Bradley S.
A1 Bowers, Aaron
A1 Tritz, Daniel
A1 Vassar, Benjamin Matthew
YR 2019
UL http://biorxiv.org/content/early/2019/09/25/773945.abstract
AB Take Home Message Many components of transparency and reproducibility are lacking in urology publications, making study replication, at best, difficult.Introduction Reproducibility is essential for the integrity of scientific research. Reproducibility is measured by the ability of investigators to replicate the outcomes of an original publication by using the same materials and procedures.Methods We sampled 300 publications in the field of urology for assessment of multiple indicators of reproducibility, including material availability, raw data availability, analysis script availability, pre-registration information, links to protocols, and whether the publication was freely available to the public. Publications were also assessed for statements about conflicts of interest and funding sources.Results Of the 300 sample publications, 171 contained empirical data and could be analyzed for reproducibility. Of the analyzed articles, 0.58% (1/171) provided links to protocols, and none of the studies provided analysis scripts. Additionally, 95.91% (164/171) did not provide accessible raw data, 97.53% (158/162) did not provide accessible materials, and 95.32% (163/171) did not state they were pre-registered.Conclusion Current urology research does not consistently provide the components needed to reproduce original studies. Collaborative efforts from investigators and journal editors are needed to improve research quality, while minimizing waste and patient risk.