Hashimoto T, Yamada M, Maeda H
Ad Hoc Committee on Statistics, Japan Pharmaceutical Manufacturers Association, Tokyo, Japan.
Nihon Yakurigaku Zasshi. 2000 Jul;116(1):23-8. doi: 10.1254/fpj.116.23.
In order to develop a new drug with scientific rationale, it is important to collect and analyze data in accordance with the protocol and to report it correctly. We suggested appropriate statistical considerations with regards to the following five points for preparing a study report on a pharmacological study: 1. For describing statistical analysis, it is important to describe the applied statistical methods concretely; 2. For describing subjects included in the analysis, it is necessary to describe the number of subjects excluded from analysis along with the reasons for their exclusion, and it is important to describe the comparability among groups; 3. For preparation of tables and figures, it is necessary to prepare tables and figures with sufficient statistical information; 4. For proper expression of the results and a meaningful discussion, we suggested the importance of drawing pharmacological conclusions with sufficient knowledge of the applied statistical methods and also suggested the statistical considerations for utilizing information obtained from the results; 5. For compiling an integrated report from two or more study reports, we suggested statistical considerations for discussing the consolidated results. We expect researchers will be able to obtain more reliable conclusions by preparing a study report taking our suggestions into consideration.
为了基于科学依据研发新药，按照方案收集和分析数据并正确报告数据非常重要。我们针对撰写药理学研究报告提出了关于以下五点的适当统计学考量：1. 对于描述统计分析，具体描述所应用的统计方法很重要；2. 对于描述纳入分析的受试者，有必要描述从分析中排除的受试者数量及其排除原因，并且描述组间可比性很重要；3. 对于表格和图表的编制，有必要编制包含足够统计信息的表格和图表；4. 为了正确表达结果并进行有意义的讨论，我们提出在充分了解所应用统计方法的情况下得出药理学结论的重要性，还提出了利用从结果中获得的信息的统计学考量；5. 对于从两份或更多研究报告编制综合报告，我们提出了讨论合并结果的统计学考量。我们期望研究人员通过考虑我们的建议来撰写研究报告，从而能够得出更可靠的结论。
