论文部分内容阅读
Bland-Altman(简称B-A)方法常用于评价两种测量方法的一致性,该方法通过绘制两种测量方法差值对均值图,并根据观察到的两种测量方法差值计算一致性界限,综合B-A图散布趋势和一致性界限的宽度,对两种方法的一致性程度进行评价。但在实际应用B-A方法的过程中,B-A图散布趋势与两种测量方法的标准差水平相关联,如果方法使用不当将错误的显示出实际并不存在的散布趋势,或者无法发现真实存在的散布趋势从而造成错判,本文将结合具体分析和随机模拟举例对上述问题加以讨论和证明。
Bland-Altman (BA) method is often used to evaluate the consistency of the two measurement methods. The method draws the difference between the two measurement methods to the mean value graph, and calculates the consistency limit according to the observed difference between the two measurement methods. The breadth of trends and the boundaries of consistency in the BA charts assess the degree of agreement between the two approaches. However, during the practical application of the BA method, the trend of BA distribution is related to the standard deviation of the two measurement methods. If the method is improperly used, the method may incorrectly display the actual non-existing dispersion trend or find the real dispersion Trend resulting in misjudgment, this article will be combined with specific analysis and stochastic simulation for example to discuss and prove the above issues.