CAJ | 학술논문

目的介绍一种临床试验中的适应性样本量调整方法,并探讨样本量调整后统计分析方法的第Ⅰ类错误率及检验效能.方法通过monte carlo模拟的方法研究n_1大小对最终样本量N_f的影响,并估计最终方差偏移大小;同时模拟研究样本量调整后统计分析方法的第Ⅰ类错误率及检验效能大小.结果 (1)模拟结果显示运用该样本量调整方法所得到的最终样本量N_f非常接近其真实值N_0,尤其在π=0.4时进行样本量调整.(2)同时模拟结果显示所介绍的样本量调整后的校正t检验方法不仅能有效控制第Ⅰ类错误率α并且能充分满足试验检验效能(1-β).结论该样本量调整方法研究结果是在一般两样本单侧t检验条件下得到也可应用于优效或非劣效设计的临床试验中.
목적 개소일충림상시험중적괄응성양본량조정방법,병탐토양본량조정후통계분석방법적제Ⅰ류착오솔급검험효능.방법 통과monte carlo모의적방법연구n_1대소대최종양본량N_f적영향,병고계최종방차편이대소;동시모의연구양본량조정후통계분석방법적제Ⅰ류착오솔급검험효능대소.결과 (1)모의결과현시운용해양본량조정방법소득도적최종양본량N_f비상접근기진실치N_0,우기재π=0.4시진행양본량조정.(2)동시모의결과현시소개소적양본량조정후적교정t검험방법불부능유효공제제Ⅰ류착오솔α병차능충분만족시험검험효능(1-β).결론 해양본량조정방법연구결과시재일반량양본단측t검험조건하득도야가응용우우효혹비렬효설계적림상시험중.
Objective Introducing a method of sample size re-estimation in adaptive clinical trials,and studying the type I error rate and the power of statistical mehod after the sample size adjustment.Methods We studied the impact of magnitude of n_1 to final sample size,assessing the bias of finally calculated variance to true variance;meanwhile studying the type I error rate and the power of the adjustment statistical mehod by monte carlo simulations.Results (1)The simulation results showed that the final sample size was very near to the true size,especially when the adjustment was made at π=0.4.(2)Meanwhile our results indicated that the bias in the type I error rate was negligible and the desired power was maintained with the introduced corrected t-test after the sample size adjustment.Conclusion All simulation results are got under the two-sample t-test with one-side and it can be applied in non-inferiority or superiority clinical trials.