CLSI Publishes Guideline for Evaluation of Quantitative Clinical Laboratory Measurement Procedures.

WAYNE, Pa. -- Before starting a complete evaluation of a new measurement procedure, kit, or instrument for in vitro diagnostic use, it is often necessary to make a preliminary decision about its acceptability. This initial performance check is neither a rigorous characterization of long-term performance nor an evaluation of the many factors that can affect results produced by the device. Instead, it is a quick check to rule out major problems, and can serve as a starting point for accumulating data and experience that will enable the user to make a final decision.

Clinical and Laboratory Standards Institute (CLSI, formerly NCCLS) recently published a document, Preliminary Evaluation of Quantitative Clinical Laboratory Measurement Procedures; Approved GuidelineCoThird Edition, which is intended to facilitate a limited, preliminary evaluation of the performance of a measurement procedure or device. Using the experimental design and data analysis procedure described, determination of whether a device has problems that require further evaluation or referral to the manufacturer can be done with a minimum expenditure of time and material.

Jan S. Krouwer, PhD, President, Krouwer Consulting, explains, "Based on a protocol designed by one of the best known industrial statisticians, Cuthbert Daniel, EP10-A3 uses a unique sampling order (three levels of three controls, allowing estimation of five parameters from one procedure)Cosaving significant time and money. This revised edition provides step-by-step instructions on how to perform the analysis in Excel, and includes specific cell formulas. This easy-to-use guideline is beneficial to both clinical laboratories running a precision study and manufacturers evaluating assays during product development."

This document is a revision of EP10-A2, published in 2002, and includes the following updates:

* a figure to illustrate which error sources the EP10 protocol can detect with respect to all error sources and other CLSI Evaluation Protocols documents;

* suggested sample sizes, useful for manufacturers;

* instructions for the multiple regression calculations in Excel;

* revised references; and

* revised definitions.

For additional information on CLSI or for further information regarding this release, visit our website at http://www.clsi.org or call +610.688.0100.

CLSI is a global, nonprofit, membership-based organization dedicated to developing standards and guidelines for the healthcare and medical testing community. CLSI's unique consensus process facilitates the creation of standards and guidelines that are reliable, practical, and achievable for an effective quality system.