Performance verification of precision, trueness, and accuracy of self-developed detection system in clinical chemistry

Objective To investigate the precision, trueness, and accuracy of self-developed detection system in clinical chemistry. Methods This was a methodological evaluation. Take serum creatine kinase(CK)for instance, 6 serum sampleswith different leves(on the upper or lower limit of the reference range or close to the medicine decide levels (MDLs), were collected for within-run precision(repeatability)and within-laboratory precision (intermediate precision)experiments. 5 proficiency testing(PT)samples, 5 samples assigned value by reference method, and 40 fresh-frozen serums were measured and compared with reference method for trueness verification. Drawing method evaluation decision chart, calculating total errors and sigma level evaluation experiment based on the CV, bias, and allowed total errors(TEa)were used to evaluate the accuracy performance.The precision, trueness, and accuracy were compared with the quality indicators. Results The within-run precision and within-laboratory precision were less than the highest requirement of Chinese industrial standard. The mean bias was -8.96%, didn′t reachthe required standard (5.5%). After taking corrective actions, all samples but one(-5.8%)met the required standard.Compared with the reference method, the mean bias on the MDLs was less than TEa. The performance points of the method evaluation decision chart indicated excellence performance. The total errors on MDLs were 14.2%, 10.4% and 7.6%, less than 15%. The sigma levels on MDLs were 5.9, 7.5 and 15, also achieved excellent level. Conclusions The precision, trueness, and accuracy performance of CK measured by self-developed detection system achieved excellent level of the Chinese industry standard, and the same results were found from different evaluation methods. (Chin J Lab Med, 2016, 39: 715-719) Key words: Precision; Trueness; Accuracy; Measuring system; Methodological evaluation