A Toxicologist's Guide to Studying Diagnostic Tests

Abstract

Diagnostic testing plays an important role in medicine, including in clinical toxicology. In comparison with therapeutics however, until recently, much less focus has been input in this vital area of research and practice. In the absence of stringent and constructive criteria, the quality and interpretability of studies evaluating diagnostic accuracy has been less than optimal. Recently, a group of researchers and clinicians have attempted to outline a checklist intended to provide guidelines for undertaking and reporting studies examining diagnostic accuracy. Also, a tool to assess the quality of diagnostic accuracy studies was developed. The purpose of this review is to inform practitioners and researchers in the field of clinical toxicology on the basic concepts underlying measurements of the accuracy of diagnostic tests, methodological concerns related to design and analysis of studies, the strengths and weaknesses of different measures of accuracy and an overview of the Standards for Reporting Diagnostic Accuracy (STARD statement) and the Quality Assessment of Diagnostic Accuracy Studies (QUADAS) tool. Understanding key concepts in the field of diagnostic testing can not only help improve future studies but also explain why diagnostic tests currently used by clinical toxicologists are not performing well in particular circumstances.

• Sensibility and specificity
• Predictive value of tests
• ROC curve
• Reproducibility of results
• Bias
• Toxicology