Hosogaya Shigemi, Ozaki Yukio
Clinical Laboratory, University of Yamanashi Hospital, Nakakoma-gun, Yamanashi-pref. 409-3898.
Rinsho Byori. 2005 Jun;53(6):547-53.
Many external quality assessment schemes (EQAS) are performed to support quality improvement of the services provided by participating laboratories for the benefits of patients. The EQAS organizer shall be responsible for ensuring that the method of evaluation is appropriate for maintenance of the credibility of the schemes. Procedures to evaluate each participating laboratory are gradually being standardized. In most cases of EQAS, the peer group mean is used as a target of accuracy, and the peer group standard deviation is used as a criterion for inter-laboratory variation. On the other hand, Fraser CG, et al. proposed desirable quality specifications for any imprecision and inaccuracies, which were derived from inter- and intra-biologic variations. We also proposed allowable limits of analytical error, being less than one-half of the average intra-individual variation for evaluation of imprecision, and less than one-quarter of the inter- plus intra-individual variation for evaluation of inaccuracy. When expressed in coefficient of variation terms, these allowable limits may be applied at a wide range of levels of quantity.
许多外部质量评估计划（EQAS）旨在支持参与实验室所提供服务的质量改进，以造福患者。EQAS组织者应负责确保评估方法适用于维护该计划的可信度。评估每个参与实验室的程序正逐渐标准化。在大多数EQAS案例中，同组均值用作准确性目标，同组标准差用作实验室间变异的标准。另一方面，弗雷泽·CG等人提出了针对任何不精密度和不准确性的理想质量规范，这些规范源自生物间和生物内变异。我们还提出了分析误差的允许限度，即评估不精密度时小于个体内平均变异的一半，评估不准确性时小于个体间加个体内变异的四分之一。以变异系数表示时，这些允许限度可应用于广泛的数量水平。
