A practical guide to validation and verification of analytical methods in the clinical laboratory

Adv Clin Chem. 2019:90:215-281. doi: 10.1016/bs.acc.2019.01.006. Epub 2019 Mar 9.

Abstract

Although the measures to improve quality in the clinical laboratory have been enormous in the past years, not least of all due to the introduction of the ISO standards 15189 and 17025, the handling of validation and verification of method performance often still differs widely from laboratory to laboratory. Much of what is published on the topic contains complex statistics and is difficult to implement in routine laboratories. The result is, that this point is often neglected or implemented incorrectly, which in turn can lead to false conclusions about method performances, potentially compromising patient safety or contributing to incorrect diagnoses. As it has long become a standard requirement for accredited laboratories to evaluate and document the analytical performance of all methods not only prior to their first implementation, but also during ongoing operation, there is a need for clear, standardized and practical guidelines on the subject. This review summarizes the current literature on the topic, focusing on the requirements for method validations, or as the case may be, verifications and describes when to validate, when to verify and which statistical tests are appropriate for each. Proper interpretation of statistical test results and acceptance criteria for each procedure are alluded to. Specific topics, which are addressed, are precision and bias verification of quantitative, qualitative and semi-quantitative procedures, method comparisons with Bland-Altman Plots, Passing-Bablok regression analysis, 2×2 contingency tables and bubble charts, linearity studies, analytical sensitivity and specificity, performing carry-over studies and establishing and confirming reference ranges.

Keywords: Bland-Altman plots; Carryover testing; Linearity studies; Method comparison; Method validation; Passing-Bablok regression; Performance verification; Precision testing; Verification of reference ranges.

Publication types

  • Review

MeSH terms

  • Clinical Laboratory Techniques / standards*
  • Humans
  • Reproducibility of Results
  • Validation Studies as Topic*