CLSI EP27
How to Construct and Interpret an Error Grid for Diagnostic Assays; Proposed Guideline
| Organization: | CLSI |
| Publication Date: | 1 July 2009 |
| Status: | inactive |
| Page Count: | 62 |
scope:
Although there are many protocols for evaluating diagnostic assays, one can always ask the question, "is the estimated performance 'good enough'"? This is a difficult question to answer, and previous attempts from CLSI and the International Organization for Standardization Technical Committee (ISO/TC) 212 to establish analytical performance goals for assays failed to be completed. This project provides a different approach by leveraging the long-standing, existing glucose error grids.1,2 EP27 explains how to construct error grids for any diagnostic assay, with a focus on the following:
• the region that should include most (95%) of the data - the allowable total error (ATE) region; and
• the region that should include ideally none (0%) of the data - the limits of erroneous results (LER) region.
Moreover, these grids are illustrated for different diseases and uses (eg, screening vs monitoring). This can lead to different error grids for the same assay. Regulatory practices, which can be different in different countries, can also affect error grids.
These grids are graphed in a spreadsheet so the output of existing evaluation protocol procedures can display data in the error grid.
Guidance is provided on how to quantify the amount of data and its corresponding confidence interval in each error grid region.b
In addition, the document is geared toward manufacturers with respect to construction of error grids. However, it is applicable also to clinical laboratories, because data from protocols such as CLSI document EP153 or EP094 can be displayed in an error grid.
This document is intended for any quantitative assay whereby both the test and comparative methods measure the same measurand using the same units.
b Note that error grids are required in the US Food and Drug Administration (FDA) guidance document for Clinical Laboratory Improvement Amendments (CLIA) waiver applications. Hence, this document supports this FDA requirement. Guidance for Industry and FDA Staff: Recommendations for Clinical Laboratory Improvement Amendments of 1988 (CLIA) Waiver Applications for Manufacturers of In Vitro Diagnostic Devices. OMB control number 0910-0598. Issued January 30, 2008.
Document History