Suboptimal practices in the reporting, dissemination, and interpretation of imaging diagnostic accuracy research
Suboptimal reporting practices, imperfect editorial processes and flawed interpretation by end users can all contribute to misrepresentation of imaging test accuracy. The objectives of this thesis were to evaluate the practices involved in the execution, reporting, and dissemination of imaging diagnostic accuracy research and to provide guidance for the interpretation of systematic reviews and meta-analyses on this topic. Chapter 2 explores publication bias in imaging research by evaluating whether conference abstracts with positive conclusions are more likely to reach full-text publication. Chapter 3 focuses on the editorial level to establish whether accuracy estimates, conclusion positivity, and abstract reporting completeness are associated with acceptance of submissions to radiology conferences and journals. Chapter 4 explores whether the peer review process improves completeness and transparency of reporting among accepted manuscripts. Chapter 5 explores citation bias in imaging research by determining whether studies with higher accuracy estimates are cited more frequently. Chapter 6 describes an evaluation of whether journal-level and study-level variables are associated with proximity of primary study results to summary estimates from meta-analyses. Chapter 7 is a systematic review evaluating the accuracy of MRI for solid renal mass diagnosis, performed using methodologic guidance from the Cochrane group and reported according to the PRISMA-DTA guideline. Chapter 8 is a guidance paper for critical appraisal and clinical application of imaging diagnostic accuracy systematic reviews. Chapter 9 integrates the findings of the preceding chapters, emphasizing potential implications and proposed solutions to optimize the reporting, dissemination, and interpretation of imaging diagnostic accuracy research.