CLSI EP21-Ed3 replaces CLSI EP21-Ed2, published in 2016. Several changes were made in this edition, including:
• Harmonizing terminology, abbreviations, and definitions related to TAE.
• Extracting content on determining allowable total error limits from Chapter 2 of the previous edition to be included in a new CLSI EP46 because the setting of acceptance criteria for allowable total error is foundational to (and as such, referenced by) several CLSI method evaluation documents.
• Additional considerations for developers (both manufacturers and laboratories that create laboratory-developed tests) for a more rigorous experimental design.
• Updating and expanding worked examples in the appendix, including the addition of a worked example with subintervals.

CLSI EP21 provides guidance for understanding, estimating, and evaluating the acceptability of total analytical error (TAE) for quantitative medical laboratory measurement procedures. This guidance is suitable for both commercially manufactured products as well as laboratory-developed tests (LDTs). It is also useful for medical laboratories to assess the performance of measurement procedures intended to be put into use. Allowable total error (ATE) limits should be determined before CLSI EP21 is used to estimate and assess TAE. Users should consult CLSI EP46 for guidance for setting limits for ATE, with consideration to how test results are used to inform patient care.
 

The intended users of CLSI EP21 are developers of measurement procedures (both commercial manufacturers and laboratories with LDTs), regulatory authorities, and medical laboratory personnel. 

Through CLSI EP21, users learn how to:
• Describe the difference between TAE and total error, which includes pre- and postanalytical components (sometimes referred to as pre- and postexamination, respectively), and understand why CLSI EP21 focuses on the former.
• Define subintervals within a measurement procedure’s analytical measuring interval (AMI) as it relates to medical decision levels affecting clinical decisions and patient management.
• Design an experiment to estimate TAE and determine if performance goals were met for the entire AMI of a measurement procedure or, when applicable, subintervals of the AMI.
 

The focus of CLSI EP21 is primarily on estimating errors occurring during the analytical phase of the testing process (total analytical error) and not the total testing process (total error), which encompasses the preanalytical through the postanalytical phases. The protocol provided estimates the combined impact of multiple sources of error, including but not limited to bias, imprecision, and other factors (eg, nonlinearity, interferences, specimento- specimen matrix differences). Although bias and imprecision both contribute to the TAE estimated using the protocol in CLSI EP21, separate estimates of these parameters are not obtained. To evaluate the bias between 2 measurement procedures, see CLSI EP09. To evaluate the precision performance of quantitative measurement procedures, see CLSI EP05. For user verification of precision and estimation of bias, see CLSI EP15. CLSI EP21 is not intended to provide guidance on identifying the source(s) of error or mitigation of TAE that falls outside of pre-established acceptance criteria. It does not cover the considerations for or approaches to setting ATE limits, but instead directs users to CLSI EP46 for information on this topic. CLSI EP21 is not intended for use in evaluating qualitative medical laboratory measurement procedures.