新药临床试验的原理与操作(6)——统计学处理的标准操作规程指南(续1)

Principle and manipulation of clinical trials(6) ——Guidelines for standard operating procedure (continued 1)

介绍制药业统计学家协会标准操作规程指南的第 5～ 8部分。强调必须仔细设计随机化编码表的产生、分发和破盲的程序 ,同时应核查是否已遵循了所制定的随机化。数据管理要有详细计划 ,特别是数据库设计、管理和维护的程序和责任人。期中分析必须事先计划 ,要尽可能避免期中分析对试验的影响。

Part 5～8 of Guidelines for standard operating procedures produced by the Statisticians in the Pharmaceutical Industry (PSI) are introduced. Thoughtful procedures for designing, distributing and breaking randomized code lists are emphasized as well as supervising the actualization of the randomization procedure worked out. There should be detailed data management plan indicating the procedure of and the one responsible for the design and maintenance of the database. Interim analysis must be aforethought and consideration should be given to avoid its possible biased impact on the running trial. The writing and contents of statistical reports are also discussed. [