Integrity of Clinical Neuroradiological Research.

Abstract

It is unclear if undesired practices such as scientific fraud, publication bias, and honorary authorship are present in neuroradiology. Therefore, the objective was to explore the integrity of clinical neuroradiological research using a survey method. Corresponding authors who published in one of four top clinical neuroradiology journals were invited to complete asurvey about integrity in clinical neuroradiology research. A total of 232 corresponding authors participated in our survey. Confidence in the integrity of published scientific work in clinical neuroradiology (0-10point scale) was rated as amedian score of 8 (range 3-10). In linear regression analysis, respondents from Asia had significantly higher confidence (beta coefficient of 0.569, 95% confidence interval, CI: 0.049-1.088, P = 0.032). Of the respondents 8(3.4%) reported to have committed scientific fraud in the past 5years, whereas 66respondents (28.4%) reported to have witnessed or suspected scientific fraud by anyone from their department in the past 5years. A total of 192 respondents (82.8%) thought that astudy with positive results is more likely to be accepted by ajournal than asimilar study with negative results and 96respondents (41.4%) had an honorary author on any of their publications in the past 5years. Experts in the field have overall high confidence in published clinical neuroradiology research; however, scientific integrity concerns are not negligible, publication bias is aproblem and honorary authorship is common. The findings from this survey may help to increase awareness and vigilance among anyone involved in clinical neuroradiological research.