The Bias Estimated and Comparative Research for two Kinds of Cholesterol Reagents / 医学研究杂志

ABSTRACT

Objective Estimating the determined result bias of two kinds of Cholesterol reagents,in order to research Comparability of results between different reagent method.Methods According to the file of NCCLS EP9-A of America,take 8 clinical serum samples every day,which conclude high、middle and low result samples,analyze each samples twice a day for 5 days,recorded results,analyzed the statistics.,to evaluate the bias estimation of two kinds reagents.According to contrast results,calibration for antomation system(c.f.a.s)taking two kinds calibration cost and using mixed human serum as calibration article,perform the same procedure described above but add 10 samples everyday.which conclude high、middle and low result samples,then statistics analysis was performed.Results The bias rate of two kinds reagents was 5.3%,the difference was statistically significant(P 0.05).Conclusions When changed regent or c.f.a.s in a clinical serum sample comparison experiment,for Cholesterol reagents from different company,c.f.a.s cannot take single calibration values,and cannot emendate the values of calibration serum only by the reply result from a single external quality assessment.Taking the mixed human serum calibrated by reference method as calibration values,which values can correct the determined bias caused by different regent.