WARNING: We do not support Internet Explorer. It is not secure and will not work correctly. Please come back using a newer web browser.


CLSI EP27

Constructing and Interpreting an Error Grid for Quantitative Measurement Procedures, 2nd Edition

This guideline provides recommendations on constructing and using error grids to evaluate the clinical acceptability of quantitative measurement procedures, based on the potential harm that may be caused by erroneous measurements with clinical consequences.

This document is available in electronic format only.

Member price:

List Price:
  PDF  Add to Cart
Log in/sign up to see your price

Details

Chairholder: Anthony Killeen, MD, PhD

Date of Publication: June 14, 2022

Order Code PDF: CLSI EP27Ed2E
ISBN Number: 978-1-68440-153-6

Order Code Print: print not available

Edition: Second

Pages: 52

CLSI EP27 Additional Details

If interested in ordering larger quantities of this document in print, please contact us here.

CLSI EP27 Abstract

Clinical and Laboratory Standards Institute guideline EP27—Constructing and Interpreting an Error Grid for Quantitative Measurement Procedures explains the function of an error grid, illustrates the concept with examples, and provides recommendations on constructing and using one. For a given measurand, error grids characterize the relationship between measurement errors and clinical management errors that may harm patients. This guideline covers the process of creating error grids with multiple error zones and describes the two primary options for doing so. After error grids have been constructed, they can be populated with data from a measurement procedure comparison experiment, as described in this guideline. A candidate measurement procedure’s clinical performance is evaluated by visually comparing the experimental data points with the error grid and tabulating the number of points that fall within each error zone. This guideline includes an example experiment that demonstrates proper interpretation of an error grid plot and its tabulated results.