Chapter 139

Quantitative Diagnostic Test Validation

First published: 04 March 2022
Citations: 1

Summary

This chapter focuses on analytical and diagnostic performance and provides an overview of the study designs and statistical tools used in their assessment. Analytical performance assessment characterizes an assay's measurement performance through the study of analytical properties such as imprecision, bias, and linearity. Individual laboratories should carry out some assessment of analytical performance prior to implementing a new assay for routine clinical use. Analytical quality goals have been derived from various sources, including regulatory requirements, expert consensus, biological variation data, “state of the art”, and patient outcome studies. Analytical performance validation of assays involving unstable specimens or labile measurands may necessarily be limited by logistical concerns. Observed total error combines observed bias and imprecision into one quantity calculated with a stated statistical probability. Diagnostic test performance validation compares results from a test under evaluation, that is, index test, to another test deemed the “gold standard” for disease diagnosis.

The full text of this article hosted at iucr.org is unavailable due to technical difficulties.