UMC Utrecht Prediction Meetings

Clinicians are often interested in knowing how the accuracy of two or more diagnostic tests compare. For example, does CT have a higher sensitivity and specificity than ultrasound for diagnosing acute appendicitis? Comparative accuracy studies assess the accuracy of multiple tests and have the potential to yield reliable evidence regarding relative test performance. However, shortcomings in the design, execution, and analysis of these studies may bias their results. QUADAS-C is a risk of bias tool for comparative accuracy studies developed through a Delphi process by 24 experts in test evaluation. It is an extension to QUADAS-2, an existing tool for assessing the quality of diagnostic accuracy studies. QUADAS-C is intended to be used in systematic reviews addressing comparative accuracy questions; it may also be used by researchers for informing the design of their study. This talk serves to introduce QUADAS-C and its key features, but we will also discuss comparative accuracy research more generally, the most important study designs, the rationale for a new risk of bias tool, and the key biases that affect comparative accuracy studies.