International Journal of Medical Biochemistry (Jan 2022)
Total analytical error assessment of Yerköy State Hospital biochemistry laboratory
Abstract
INTRODUCTION: In 2016, the maximum allowable CV% and allowable total error limits (TEa-TR) for 15 test parameters were specified by the Ministry of Health of Turkey and put into practice. Our study aimed to assess the analytical performance of 15 test parameters studied in our laboratory in terms of the TEa-TR criteria and demonstrate the effect of the method of CV% use in the TAE formula on the performance of tests when two-level internal quality control (IQC) was employed. METHODS: Total analytical error (TAE) was calculated using the formula TAE%=(CV%×1.65)+bias%. Two different procedures were followed for TAE calculation. TAE1 and TAE2 were separately calculated for each control level using the CV% values obtained from the data of IQC at two different levels used for each test. In addition, TAE3 values were found for each test parameter in line with the total CV% value (Total CV%=√(CV12+CV22) obtained using the IQC data from two different levels. RESULTS: When CV% values were separately taken for each control level, the maximum allowable CV% value was exceeded in two tests at IQC Level 2, whereas the TEa-TR limit was exceeded in two tests at IQC Level 1 and in 2 tests at IQC Level 2 during the 6-month period. According to the TEa-TR formula, the allowable CV% was exceeded 19 times in 13 out of 15 tests, and TEa limits were exceeded 15 times in 10 tests during the 6-month period DISCUSSION AND CONCLUSION: When the total CV% was calculated from Level 1 and Level 2 according to the TEa-TR formula, both the maximum allowable CV% and allowable TEa% limits were exceeded many times. However, when the CV% and TAE values were calculated separately for different levels of IQC samples, the analytical performance of the tests were found better. We think that calculating the TAE and CV% values separately is more useful in showing the analytical performance of the test at different levels.
Keywords