Moskowitz Chaya S, Pepe Margaret S
Department of Epidemiology and Biostatistics, Memorial Sloan-Kettering Cancer Center, New York, USA.
Clin Trials. 2006;3(3):272-9. doi: 10.1191/1740774506cn147oa.
Although statistical methodology is well developed for comparing diagnostic tests in terms of their sensitivities and specificities, comparative inference about predictive values is not.
In this paper we consider the design and analysis of studies comparing the positive and negative predictive values of two diagnostic tests that are measured on all subjects.
We focus on comparing tests using the relative positive and negative predictive values. We discuss directly estimating these quantities from the data and derive analytic variance expressions. Sample size formulas for study design ensue.
We analyze data on patients with cystic fibrosis to illustrate the methodology. This approach is compared and contrasted with an existing regression framework that can also be used for similar analysis purposes and yields similar results.
We have developed a new approach for comparing the predictive values of two tests that gives rise to sample size formulas for study design.
尽管在比较诊断试验的敏感性和特异性方面，统计方法已经很成熟，但对于预测值的比较推断却并非如此。
在本文中，我们考虑对在所有受试者身上测量的两种诊断试验的阳性和阴性预测值进行比较的研究的设计与分析。
我们专注于使用相对阳性和阴性预测值来比较试验。我们讨论直接从数据中估计这些量，并推导分析方差表达式。由此得出研究设计的样本量公式。
我们分析了囊性纤维化患者的数据以说明该方法。将此方法与现有的回归框架进行了比较和对比，该回归框架也可用于类似的分析目的并产生类似的结果。
我们开发了一种比较两种试验预测值的新方法，该方法产生了研究设计的样本量公式。
