Centre for Experimental Cancer Medicine, Barts Cancer Institute, Queen Mary, University of London, UK.
Comput Methods Programs Biomed. 2013 Jun;110(3):471-89. doi: 10.1016/j.cmpb.2013.01.008. Epub 2013 Feb 8.
Cost-Effectiveness Analysis (CEA) has become an increasingly important component of clinical trials. However, formal sample size calculations for such studies are not common. One of the reasons for this might be due to the absence of readily available computer software to perform complex calculations, particularly in a Bayesian setting. In this paper, a C++ program (using NAG library functions/subroutines) is presented to estimate the sample sizes for cost-effectiveness clinical trials in a Bayesian framework. The program can equally be used to calculate sample sizes for efficacy trials. The Bayesian approach to sample size calculation is based on that of O'Hagan and Stevens (A. O'Hagan, J.W. Stevens, Bayesian assessment of sample size for clinical trials of cost-effectiveness, Medical Decision Making 21 (2001) 219-230). With this program, the user can calculate sample sizes for various thresholds of willingness to pay and under various assumptions of the correlations between cost and effects. Under some prior, the program produces frequentist sample size as well. The program runs under windows environment and running time is very short.
成本效益分析（CEA）已成为临床试验中越来越重要的组成部分。然而，此类研究的正式样本量计算并不常见。造成这种情况的原因之一可能是缺乏可用于执行复杂计算的现成计算机软件，特别是在贝叶斯环境中。本文介绍了一个 C++程序（使用 NAG 库函数/子程序），用于在贝叶斯框架下估算成本效益临床试验的样本量。该程序同样可用于计算疗效试验的样本量。基于 O'Hagan 和 Stevens 的贝叶斯样本量计算方法（A. O'Hagan，J.W. Stevens，用于成本效益临床试验样本量评估的贝叶斯方法，Medical Decision Making 21 (2001) 219-230）。使用此程序，用户可以根据不同的意愿支付阈值和成本与效果之间的相关性的不同假设来计算样本量。在某些先验条件下，程序还会生成频率样本量。该程序在 Windows 环境下运行，运行时间非常短。
