Evaluation of 20 clinical chemistry and 12 immunoassay analytes in terms of total analytical error and measurement uncertainty.

In analytical quality management, target setting models that are selected by the purpose together with the error models that are applied correctly have critical importance. In our study, we aimed to compare the analytical performance characteristics of routine clinical chemistry and immunoassay tests with different target-setting models proposed by various organizations. Our study was performed with internal and external quality control data obtained using Beckman Coulter AU680 for clinical chemistry analytes and Roche Cobas 8000 autoanalyzer for immunoassay analytes. The total analytical error (TAE) was calculated by the formula TAH%=1.65×(CV%)+Bias%. Measurement uncertainty (MU) has been calculated adhering to the Nordtest guideline. Results were compared with BVEFLM, CLIA, RCPA, PRDEQA%, pUEQAS%, and permissible MU (pU%) data to investigate analytical performance qualities. When we compare the results of TAE and MU, MU was found to be higher than TAE for all analytes. ALT, AST, glucose, K, and triglycerides met all target values, showing the best performance. Taken together, our results show that CLIA for total analytical error and PRDEQA% and pUEQAS% for measurement uncertainty can match better than BVEFLM, RCPA and pU%. These test results should be evaluated with measurement uncertainty to avoid misdiagnosis. Appropriate specification limits should be defined for the examination of test methods that meet the objectives for fitness for clinical purposes.