中国卒中杂志 ›› 2022, Vol. 17 ›› Issue (01): 31-35.DOI: 10.3969/j.issn.1673-5765.2022.01.004

• 专题论坛 • 上一篇    下一篇

临床研究样本量的估计方法和常见错误

潘岳松, 金奥铭, 王梦星   

  1. 北京 100070首都医科大学附属北京天坛医院,国家神经系统疾病临床医学研究中心
  • 收稿日期:2021-10-18 出版日期:2022-01-20 发布日期:2022-01-20
  • 通讯作者: 潘岳松 panys2000@163.com

Methods and Common Pitfalls of Sample Size Estimation in Clinical Studies

  • Received:2021-10-18 Online:2022-01-20 Published:2022-01-20

摘要: 合理的样本量估计是临床研究设计的重要环节,也是临床医师进行科研设计需关注的重 要问题之一。临床研究的样本量估计与研究目的、研究设计、主要观察指标的资料类型、效应值、变 异度、检验水平和把握度等因素有关。临床研究设计中样本量估计的常见错误包括未经过计算直接 确定样本量、样本量估算方法与研究设计和主要观察指标不对应,参数设置不合理等。

关键词: 样本量; 统计功效; 临床研究; 研究设计

Abstract: Appropriate sample size estimation was an important issue in clinical research design that clinical researchers pay attention to. Study objectives, study design, data type of the primary outcome, effect of interest, variability of the outcome measurement, type I error and power of test were key issues related to estimation of sample size in clinical studies. The main pitfalls of sample size estimation in clinical research design included direct determination of sample size without calculation, incompatibility between sample size estimation method and study design and objectives, and unreasonable parameter setting, etc.

Key words: Sample size; Statistical power; Clinical research; Study design