B-124 Applicability of measurement uncertainty in the validity of results in the clinical laboratory

Abstract

Abstract Background The results of the clinical laboratory must comply with attributes of precision, veracity, and accuracy to guarantee the validity of the result; since the assurance of analytical quality, different performance evaluation strategies have been applied with the use of tools such as total permissible error (ETa) and sigma metric. However, based on the requirements to qualify for the accreditation process by standards such as ISO 15189 and ISO 17025, the concept of the “measurement uncertainty” of the measured magnitude values was introduced, giving special relevance in the interpretation of the values obtained, therefore it is important to consider its estimation and evaluation as a tool that contributes to patient safety in the interpretation of clinical decision limits. Methods Uncertainty estimation was performed for forty-one measurands that were previously selected considering the variables of clinical risk and processing volume. The estimation of uncertainty, ETa and sigma metric were made with the RANDOX Acusera software 24/7 online from the information obtained from the internal quality control data and the scheme of the external quality evaluation. The uncertainty obtained was compared against four criteria of acceptability of uncertainty. Results After comparing the uncertainty obtained against the four criteria of acceptability of uncertainty, it was found that the most demanding criterion is that derived from the reference biological interval. The implementation of this criterion would allow objective comparability of performance between laboratories. Conclusions: The estimation of uncertainty as an indicator of the validity of the results provides relevant information in the interpretation of the results that are within the limits of clinical decision, while the ET and sigma metric contribute to the monitoring and control of the performance of the analysis or measurement procedure. Conclusion We observed that the efforts of the laboratory for minimizing the variation sources that can affect the analytical accuracy have been prolific. Total Error quantification, Sigma 6 and MU have a correlation, favorably impacting the tests performance. Aiming to guarantee a constant improvement of the results’ quality, the Total Error and MU goals must be periodically revised, ensuring their utility throughout time.