率的单组目标值试验样本量估计参数设置模拟研究  

A single-group target value test sample size estimation parameter setting simulation study

作　　者：冯玉婷 陶立元[2] 柴倩云[1] 罗慜婧 高一城 贾金柱 费宇彤[1] FENG Yuting;TAO Liyuan;CHAI Qianyun;LUO Minjing;GAO Yicheng;JIA Jinzhu;FEI Yutong(Center for Evidence-based Medicine,Beijing University of Chinese Medicine,Beijing 100029,P.R.China;Research Center of Clinical Epidemiology,Peking University Third Hospital,Beijing 100191,P.R.China;School of Public Health and Centerfor Statistical Science,Peking University,Beijing 100080,P.R.China)

机构地区：[1]北京中医药大学循证医学中心,北京100029 [2]北京大学第三医院临床流行病研究中心,北京100191 [3]北京大学公共卫生学院和统计科学中心,北京100080

出　　处：《中国循证医学杂志》2022年第9期1062-1070,共9页Chinese Journal of Evidence-based Medicine

基　　金：国家自然科学基金项目(编号:82074282)。

摘　　要：目的针对以率为结局评价指标的单组目标值临床试验,探讨不同样本量估算方法的参数设置和估算结果间的差异大小。方法通过文献回顾探讨单组目标值临床试验的目标值选择方法,通过数值模拟设定不同取值的目标值(P_(0))、临床预期值(P_(1))和Ⅱ类错误水平(β),通过PASS软件得出不同样本量估算方法下的样本量估算结果,采用变异系数、极差/均数、方差分析等方法比较不同样本量估算方法间的差异。结果对数据模拟结果进行分析显示:(1)当预期值P_(1)固定不变时,在样本量极限值两侧随着目标值P_(0)的增加或减少,样本量的下降速度由快变慢。(2)当P_(0)与P_(1)相差在0.15范围内时,校正前后比值可控制在0.9以内;当P_(0)与P_(1)相差在0.6以上时,校正前后比值可逼近0.5。(3)当P_(0)+P_(1)≈1时,不同标准误的选择(Sp_(0)或Sp_(1))对样本量估算结果的比值接近1;当0.65<P_(0)+P_(1)<1.35时,不同标准误的选择(Sp_(0)或Sp_(1))对样本量估算结果的比值在3倍左右。(4)当把握度为0.8时,P_(0)和P_(1)的取值分别在0.25~0.75、0.20~0.80时,五种方法估算的样本量结果的差异较小(CV<0.10,极差/均数<0.2)。结论不同样本量估算方法间存在一定差异,但当P_(0)和P_(1)的取值均在0.5附近时不同方法间差异较小,提示在样本量估算时需要选择恰当的方法。Objective To explore the parameter selection of different sample size estimation methods and the differences in estimation results in single-group target value clinical trials with rate as the outcome evaluation index.Methods We conducted a literature review to assess the method of target value selection for single-group target value clinical trials.Then,different values of target value(P_(0)),clinical expected value(P_(1)),and class II error level(β)were set through numerical simulation.Sample size results estimated using different sample size estimation methods were obtained using PASS software.The coefficient of variation,range/mean,analysis of variance and other methods were used to compare the differences between different methods.Results Analysis of the data simulation results showed:when the expected value P_(1) was fixed,the sample size first decreased rapidly and then decreased slowly along with the increase or decrease of the targeted value P_(0) on both sides of the sample size limit value.When the difference between P_(0) and P_(1) was within 0.15,the ratio before and after correction could be controlled within 0.9.When the difference between P_(0) and P_(1)was more than 0.6,the ratio before and after correction approached 0.5.When P_(0)+P_(1)≈1,the ratio of different standard error choices(Sp_(0) or Sp_(1))to the estimated sample size was close to 1.When 0.65<P_(0)+P_(1)<1.35,the ratio of different standard errorchoices(Sp_(0) or Sp_(1))to the estimated sample size was about 3:1.When the confidence was 0.8,P_(0)and P_(1) were between 0.25 and 0.75 and between 0.20 and 0.80,respectively.We found little difference among the sample sizes estimated using these five methods(CV<0.10,range/mean<0.2).Conclusion There are some differences among different sample size estimation methods,however,when P_(0)and P_(1) values are around 0.5,the dfferences between different methods are small,suggesting that appropriate methods should be selected for sample ize estimation.

关 键 词：单组试验 目标值法 样本量估算 连续性校正 

分 类 号：R195.1[医药卫生—卫生统计学]