Knowles R C, Gilmore B F
Am J Clin Pathol. 1981 Oct;76(4 Suppl):590-6.
During the period from January 1979 through January 1981, 155 active participants in the Microbiology Program of the College of American Pathologists Quality Assurance Service (QAS) submitted approximately two million individuals determinations on three quality control reference strains. Of these determinations, 83% were obtained using the standardized disc-diffusion procedure of Bauer and co-workers, and, 6% using the agar overlay modification of Barry and associates. The mean and standard deviation for data obtained using these procedures were similar to those for data collected from the inception of the program in 1974 through December 1978. The number of individual determinations falling above or below existing control limits, however, was significantly less for the present data than for that previously reported. These differences result from a change in computation procedures. In the current report percentages of daily values exceeding the NCCLS limits were calculated using actual counts, whereas in the previous reports these percentages were estimated by accepting an assumption of Gaussian distribution.
在1979年1月至1981年1月期间,美国病理学家学会质量保证服务(QAS)微生物学项目的155名活跃参与者对三种质量控制参考菌株进行了约两百万次个体检测。在这些检测中,83%是使用鲍尔及其同事的标准化纸片扩散法获得的,6%是使用巴里及其同事的琼脂覆盖改良法获得的。使用这些方法获得的数据的均值和标准差与1974年该项目启动至1978年12月收集的数据相似。然而,目前数据中高于或低于现有控制限的个体检测数量明显少于先前报告的数据。这些差异是由计算程序的变化导致的。在本报告中,超过NCCLS限值的每日值百分比是使用实际计数计算的,而在先前的报告中,这些百分比是通过接受高斯分布假设来估计的。