两种测量结果的一致性和差异性评价方法

蒋伏松1, 2

（1.上海交通大学附属第六人民医院内分泌代谢科，上海，200233；2.上海健康医学院附属第六人民医院东院 内分泌代谢科，上海，201306）

浏览次数：611次 下载次数：421次

摘要：

在临床和科研工作中，有时需要评价两种仪器、两种方法或两个测量者测量结果的一致性和差异性。观察指标可分为计数资料和计量资料，常用的评价方法有卡方检验、Kappa 检验、相关分析、组内相关系数、Passing-Bablok 回归、配对t检验、 Bland-Altman 散点图、误差分析图等。选择合适的评价方法非常重要。本文拟就评价两种仪器、方法、测量者的检测结果的一 致性和差异性的方法作简单概括，以供临床医师参考。

关键词：评价；一致性；差异性

中图分类号：R446文献标志码：A文章编号：2096-1413(2017)28-0116-02

Consistency and variability evaluation of two kinds of measurement results
JIANG Fu-song 1, 2
(1. Department of Endocrine and Metabolic, Sixth People``s Hospital Affiliated to Shanghai Jiao Tong University, Shanghai 200233; 2. Department of Endocrine and Metabolic, East Hospital of Sixth People``s Hospital Affiliated to Shanghai University of
Medicine & Health Sciences, Shanghai 201306, China)

ABSTRACT: In clinical and scientific research, sometimes it is necessary to evaluate the consistency and variability of the results of two instruments, two methods, or two surveyors. Observation indicators can be divided into counting data and measurement data. The methods commonly used were chi -square test, Kappa test, correlation analysis, intra -group correlation coefficient, Passing -Bablok regression, paired t test, Bland -Altman scatter plot, error analysis, et al. It is
important to choose the appropriate evaluation method. Thus this paper intended to provide a brief overview of the methods used to evaluate the consistency and variability of two instruments, two methods, or two surveyors, to provide reforence for clinicians.
KEYWORDS: evaluation; consistency; variability

参考文献：
[1] 夏邦世,吴金华.Kappa 一致性检验在检验医学研究中的应用[J].中 华检验医学杂志,2006,29(1):83-84.
[2] ATIQI R,VAN IC,CLEOPHAS TJ.Accuracy assessments of quantitative diagnostic tests for clinical research [J].Int J Clin Pharmacol Ther, 2009,47(3):153-158.
[3] BILI ZULLE L.Comparison of methods: Passing and Bablok regression.[J].Biochem Med,2011,21(1):49-52.
[4] CUTHBERT SC,JR GG.On the reliability and validity of manual muscle testing: a literature review[J].Chiropr Osteopat,2007,15(1):4.
[5] BUNCE C.Correlation, agreement, and Bland-Altman analysis: statistical analysis of method comparison studies[J].Am J Ophthal,2009,148 (1):4-6.
[6] ZAKI R,BULGIBA A,ISMAIL NA.Testing the agreement of medical instruments: overestimation of bias in the Bland -Altman analysis [J]. Prev Med,2013,57(3):S80-82.
[7] CLARKE WL,COX D,GONDERFREDERICK LA,et al.Evaluating clinical accuracy of systems for self-monitoring of blood glucose[J].Diabetes Care,1987,10(5):622-628.
[8] FUSONG J,XUHONG H,JIA WP,et al.Assessment of the performance of A1CNow(+) and development of an error grid analysis graph for comparative hemoglobin A1c measurements[J].Diabetes Technol Ther,2014,16 (6):363-369.