临床研究中效应估计的统计分析与统计图表

Statistical methods and statistical charts for effect estimation in clinical research

临床研究主要目的是评估某干预措施的效果,或某暴露因素与结局的关联强度,此即效应估计。依据结局变量的类型,效应估计可分为三大类:(1)连续性结局变量的均数差估计;(2)分类结局变量的率差、危险度比以及比值比估计;(3)生存结局数据的生存率和风险比估计。本文将结合研究实例阐述效应估计涉及的统计分析方法及统计图表。

The main objective of a clinical research program is to assess the effect of interventions or evaluate the association between exposures and outcomes.That’s what effect estimation means.Based on different types of outcomes variables,effect estimation can be summarized into three categories:①mean difference for continuous outcomes;②rate difference,risk ratio or odds ratio for binary outcomes;③survival rate and hazard ratio for survival data.This paper aims to provide a summary and real examples of statistical methods and statistical charts related to effect estimation in clinical research.