Evaluating radiographers' diagnostic accuracy in screen-reading mammograms: What constitutes a quality study?

Abstract

Introduction: The aim of this study was to first evaluate the quality of studies investigating the diagnostic accuracy of radiographers as mammogram screen-readers and then to develop an adapted tool for determining the quality of screen-reading studies. Methods: A literature search was used to identify relevant studies and a quality evaluation tool constructed by combining the criteria for quality of Whiting, Rutjes, Dinnes et al. and Brealey and Westwood. This constructed tool was then applied to the studies and subsequently adapted specifically for use in evaluating quality in studies investigating diagnostic accuracy of screen-readers. Results: Eleven studies were identified and the constructed tool applied to evaluate quality. This evaluation resulted in the identification of quality issues with the studies such as potential for bias, applicability of results, study conduct, reporting of the study and observer characteristics. An assessment of the applicability and relevance of the tool for this area of research resulted in adaptations to the criteria and the development of a tool specifically for evaluating diagnostic accuracy in screen-reading. Conclusions: This tool, with further refinement and rigorous validation can make a significant contribution to promoting well-designed studies in this important area of research and practice. This paper has been written firstly to evaluate the quality of studies that evaluate the accuracy of radiographers' screen-reading mammograms. Secondly, an appropriated tool has been developed to determine the essential issues of quality when evaluating these studies. We believe that this tool, with further refinement, and application can contribute to future quality studies in this important area of research and practice.